Summary
CrewAI is designed around the idea of collaborative, role-based AI agents — hence the name "Crew," highlighting teamwork. Each agent is assigned a specific role, enabling them to work together to solve complex tasks. Unlike single-agent frameworks, CrewAI supports inter-agent communication and dynamic task delegation, allowing tasks to be intelligently split and shared. In this notebook, we demonstrated how to use CrewAI agents both with and without a manager, integrated web search capabilities using LangChain and SerpAPI, and generated personalized messages based on the gathered information.
Python functions and data files needed to run this notebook are available via this link.
In [1]:
import warnings
warnings.filterwarnings('ignore')
# Core classes from the CrewAI framework used to build agent workflows
from crewai import Agent, Task, Crew, Process
import codecs
import os
# Utility tools provided by the separate 'crewai_tools' package
from crewai_tools import (
ScrapeWebsiteTool, # Extracts raw information from a given website
FileWriterTool, # Writes retrieved data to a file
TXTSearchTool # Searches for specific content within a text file
)
import requests
import os
from langchain_openai import ChatOpenAI
CrewAI¶
- CrewAI is built around the concept of collaborative, role-based AI agents—hence the name “Crew,” emphasizing teamwork.
- The core idea is that multiple agents, each assigned a specific role, can work together like a team to tackle complex tasks.
- Unlike single-agent frameworks, CrewAI enables inter-agent communication and dynamic task delegation, allowing tasks to be split and passed between agents intelligently.
🔑 Key Features of CrewAI
Role-Based Architecture:
Each agent is assigned a distinct role (e.g., researcher, planner, writer) tailored to the overall task.Multi-Agent Orchestration:
Coordinates the execution of multiple agents working collaboratively.Dynamic Task Delegation:
Tasks can be passed between agents based on their role or expertise—no need for one agent to do everything.Hierarchical & Sequential Workflows:
Supports both high-level "manager" agents and low-level "worker" agents in a structured task flow.
See Figure below for visuals CrewAI workflow:
Retrieved image from https://ai.plainenglish.io/mastering-crewai-chapter-1-your-first-intelligent-workflow-06b58aed1b3d
🤝 CrewAI’s Multi-Agent Approach
Role-Based Design
- Each agent is assigned a specific responsibility, such as research, analysis, writing, or summarization.
- This mirrors how teams operate in the real world, with each member focusing on their strength.
Autonomous Inter-Agent Delegation
- Agents can dynamically delegate tasks to one another based on the needs of the workflow.
- This enables a more flexible and scalable approach to complex tasks.
Agents with CrewAI¶
In [2]:
# Create an instance of the web scraping tool targeting CBC News homepage
cbc_news_scraper = ScrapeWebsiteTool(
website_url='https://www.cbc.ca/news'
)
# Run the scraper to extract the raw textual content from the webpage
cbc_news_content = cbc_news_scraper.run()
# Display the first 100 characters of the scraped content as a preview
print(cbc_news_content[:100])
Using Tool: Read website content CBC News - Latest Canada, World, Entertainment and Business News Content Skip to Main Content Acces
In [3]:
# Initialize the tool
file_writer_tool = FileWriterTool()
# Write content to a file in a specified directory
result = file_writer_tool.run(
filename='cbc_news.txt',
content=cbc_news_content,
directory='./output',
overwrite='true'
)
print(result)
Using Tool: File Writer Tool Content successfully written to ./output\cbc_news.txt
In [4]:
txt = '../output/cbc_news.txt'
# Open the original file with the correct source encoding
with codecs.open(txt, 'r', encoding='cp1252') as source_file:
content = source_file.read()
# Write the content to the file with UTF-8 encoding
with codecs.open(txt, 'w', encoding='utf-8') as target_file:
target_file.write(content)
In [5]:
# OpenAI key
os.environ['OPENAI_API_KEY'] = 'sk-.........'
# Set up the document search utility using a specific text file for content-based queries
# Utilizes open source ChromaDB to split and vectorize the file data
news_search_engine = TXTSearchTool(txt=txt)
D:\Learning\MyWebsite\FinalGithub\ToPublihsed\projects\Agentic_Workflow\agentic_workflow\Lib\site-packages\chromadb\types.py:144: PydanticDeprecatedSince211: Accessing the 'model_fields' attribute on the instance is deprecated. Instead, you should access this attribute from the model class. Deprecated in Pydantic V2.11 to be removed in V3.0. return self.model_fields # pydantic 2.x Inserting batches in chromadb: 100%|█████████████████████████████████████████████████████| 1/1 [00:02<00:00, 2.11s/it]
Create Agent and Task without Assigned Manager¶
In [6]:
# Define a news_interpreter agent
news_interpreter = Agent(
role='News Interpreter', # Role of the agent
goal='Analyze the provided news content and answer the related question.', # Agent's objective
backstory='You are skilled at interpreting news events, and offering opinions on the significance of the news.',
verbose=True,
allow_delegation=False, # This agent handles tasks independently and does not delegate
tools=[news_search_engine]
)
# Define a summarize_news task
summarize_news_task = Task(
description="Grasp the main idea of news events, summarize them, and share your perspective on their importance.",
agent=news_interpreter, # Assigning the task to our agent
expected_output='Provide a concise response in bullet points.'
)
In [7]:
# Create the crew with a sequential process
news_summary_crew = Crew(
tasks=[summarize_news_task],
process=Process.sequential # Task execution follows a linear, step-by-step order
)
# Run the workflow
output_news = news_summary_crew.kickoff()
# Agent: News Interpreter ## Task: Grasp the main idea of news events, summarize them, and share your perspective on their importance.
D:\Learning\MyWebsite\FinalGithub\ToPublihsed\projects\Agentic_Workflow\agentic_workflow\Lib\site-packages\chromadb\types.py:144: PydanticDeprecatedSince211: Accessing the 'model_fields' attribute on the instance is deprecated. Instead, you should access this attribute from the model class. Deprecated in Pydantic V2.11 to be removed in V3.0. return self.model_fields # pydantic 2.x
# Agent: News Interpreter ## Thought: I need to gather relevant news content to summarize and provide insights on its significance. ## Using tool: Search a txt's content ## Tool Input: "{\"search_query\": \"news events\"}" ## Tool Output: Relevant Content: CBC News - Latest Canada, World, Entertainment and Business News Content Skip to Main Content Accessibility Help Menu When search suggestions are available use up and down arrows to review and enter to select. Search Search Sign In Quick Links Currently Selected: News News Sports Radio Music Listen Live TV Watch news Currently Selected: Top Stories Top Stories Local Climate World Canada Politics Indigenous Business The National Health Entertainment Science CBC News Investigates Go Public About CBC News Being Black in Canada More CBC News - Latest Canada, World, Entertainment and Business News Loaded CBC News - Latest Canada, World, Entertainment and Business News Featured Content Liberal operatives planted 'stop the steal' buttons at conservative conference Liberal operatives planted 'stop the steal' buttons at conservative conference Conservatives accuse Liberals of ‘attempting to bring American-style politics’ to Canada Breaking 14 minutes ago First Person The oil patch gave me a great career. I want a party that sees its potential for national unity January 1, 1970 Compare party promises on top election issues January 1, 1970 How are the parties doing? Check CBC's Poll Tracker January 1, 1970 Video Singh agrees Indigenous issues have not been a major campaign focus January 1, 1970 Analysis After wrapping himself in the maple leaf, Carney needs Liberal support to stick like syrup January 1, 1970 New Poilievre attacks Carney again over ethics, Singh promises help for northern Ontarians January 1, 1970 ChatGPT let us create misleading fakes of Carney and Poilievre. It's not supposed to do that Image generator is supposed to block certain content, but ChatGPT suggested workarounds Canada | 3 World - World Photos | April 11 TRADE WAR Trump administration to exclude smartphones and computers from latest tariffs April 12 RCI is CBC/Radio-Canada's multilingual service, forging bonds between Canadians and citizens of the world visit Radio Canada International Watch CBC News Network WATCH: CBC News Network Stream CBC News Network on CBC Gem News | November 12, 2024 WATCH: CBC News Explore See what's live now on CBC News CBC News. We explain the news. Stream live News | January 8 CBC Ombudsman Feedback or complaints regarding the application of CBC Journalistic Standards and practices?   Learn more Featured Video now Featured Video - 10 videos Video 1:57 Why Montreal is cutting down some century-old trees in this park 11 hours ago Video 12:11 Smartphones, computers to be excluded from U.S. tariffs, says White House CBC News | April 12 Video 2:03 Ryan Reynolds urges renaming Ontario arena after young girl lost to cancer The National | April 12 Video 2:17 Poilievre dismisses campaign criticism from former Harper strategist The National | April 12 Video 1:29 #TheMoment a Canadian reality TV star got her stolen pug back The National | April 12 Video 0:45 Belgian chocolatiers hatch exhibition for charity April 11 Video 2:06 Will there be a national unity crisis if Conservatives lose? April 12 Video 7:32 GM plant chair says Ingersoll plant shutdown 'devastating' to community April 11 Video 9:51 Hundreds of layoffs expected at Ingersoll, Ont., assembly plant as GM halts production CBC News | April 11 Video 2:11 Why you told us the federal election feels like a big deal April 11 More CBC News Israeli missiles strike Gaza hospital shortly after warning to evacuate World | 20 minutes ago Measles outbreaks spark concern over rare 'horrific' neurological disorder April 12 Why # Agent: News Interpreter ## Final Answer: - **Liberal operatives planted 'stop the steal' buttons at conservative conference**: Conservatives have accused the Liberals of instilling American-style politics in Canada, igniting heightened political tension. - **Calls for action on Indigenous issues**: Jagmeet Singh acknowledges that Indigenous matters haven't been prioritized in the campaign, suggesting a gap in political discourse. - **Carney's appeal for Liberal support**: As he wraps himself in nationalist symbols, Carney's success hinges on garnering backing from the Liberal party, highlighting party dynamics. - **Criticism towards Poilievre regarding ethics**: New criticisms from Pierre Poilievre regarding Carney's ethics reflect ongoing party rivalries and ethical debates in politics. - **Job losses in Ingersoll plant**: Hundreds of layoffs are anticipated at the GM plant, raising concerns about community impact and economic stability. - **Missiles strike Gaza hospital**: Ongoing conflict in Israel intensifies with missile strikes hitting a hospital, raising international concern for civilian safety. - **Measles outbreaks raising alarm**: Health experts express worry over measles outbreaks, connecting them to serious neurological conditions, stressing the need for vaccination and public health measures.
There are two types of processes in Crew: sequential and hierarchical:
- Sequential is the simpler one—it processes tasks one after another in a fixed order.
- Hierarchical involves planning and delegation. Instead of following a set order, it determines the most effective sequence to complete the tasks and assigns them to the appropriate agents.
In [8]:
print(output_news.raw)
- **Liberal operatives planted 'stop the steal' buttons at conservative conference**: Conservatives have accused the Liberals of instilling American-style politics in Canada, igniting heightened political tension. - **Calls for action on Indigenous issues**: Jagmeet Singh acknowledges that Indigenous matters haven't been prioritized in the campaign, suggesting a gap in political discourse. - **Carney's appeal for Liberal support**: As he wraps himself in nationalist symbols, Carney's success hinges on garnering backing from the Liberal party, highlighting party dynamics. - **Criticism towards Poilievre regarding ethics**: New criticisms from Pierre Poilievre regarding Carney's ethics reflect ongoing party rivalries and ethical debates in politics. - **Job losses in Ingersoll plant**: Hundreds of layoffs are anticipated at the GM plant, raising concerns about community impact and economic stability. - **Missiles strike Gaza hospital**: Ongoing conflict in Israel intensifies with missile strikes hitting a hospital, raising international concern for civilian safety. - **Measles outbreaks raising alarm**: Health experts express worry over measles outbreaks, connecting them to serious neurological conditions, stressing the need for vaccination and public health measures.
Create Agent and Task with Assigned Manager¶
In [9]:
# Define a new agent with a psychology-focused role
mental_health_advisor = Agent(
role='Mental Health Advisor',
goal='Interpret the given context and provide thoughtful responses to psychological questions.',
backstory=(
'You are a licensed mental health professional with expertise in psychological methods, '
'dedicated to helping individuals understand and improve their mental and emotional well-being.'
),
verbose=True,
allow_delegation=True,
tools=[]
)
# Define a task for addressing a child’s sleep issues
child_sleep_support_task = Task(
description="How can I support my son who has a sleep disorder?",
expected_output='Provide trustworthy and actionable advice in summarized 3-5 bullets.'
)
In [10]:
# Define a translator agent
translator = Agent(
role='Translator', # Role of the agent
goal='Review the provided psychological text and translate it according to the specified question.', # Agent's objective
backstory='You are a skilled translator specializing in psychology, capable of translating English \
psychological content into French, Spanish, and Farsi.',
verbose=True,
allow_delegation=True, # This agent handles tasks independently and does delegate
)
# Define a translate_summarize task
translate_summarize_task = Task(
description="Understand the main idea and translate it into Farsi as a summarized version in 3–4 bullet points.",
expected_output='Provide a concise response into Farsi language in 3-4 bullet points.'
)
In [11]:
# Initialize the crew with tasks and agents, letting the manager decide the workflow
support_crew = Crew(
tasks=[child_sleep_support_task, translate_summarize_task],
agents=[mental_health_advisor, translator],
process=Process.hierarchical, # Allow the manager to determine task priority and delegation
manager_llm=ChatOpenAI(model_name="gpt-3.5-turbo", temperature=0.2), # Manager decide which task should be assign to which agent
verbose=True
)
crew_result = support_crew.kickoff()
╭──────────────────────────────────────────── Crew Execution Started ─────────────────────────────────────────────╮ │ │ │ Crew Execution Started │ │ Name: crew │ │ ID: a70ebb8e-4c71-4b37-9700-4aed94269a68 │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew └── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 Status: Executing Task...
🚀 Crew: crew └── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 Status: Executing Task... └── 🤖 Agent: Crew Manager Status: In Progress
# Agent: Crew Manager ## Task: How can I support my son who has a sleep disorder?
🤖 Agent: Crew Manager Status: In Progress └── 🧠 Thinking...
🤖 Agent: Crew Manager Status: In Progress
🤖 Agent: Crew Manager Status: In Progress
🚀 Crew: crew └── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 Status: Executing Task... ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Mental Health Advisor Status: In Progress
# Agent: Mental Health Advisor ## Task: Research effective ways to support a child with a sleep disorder # Agent: Mental Health Advisor ## Final Answer: Supporting a child with a sleep disorder involves a variety of strategies tailored to the specific nature of the disorder, the child's age, and individual circumstances. Here are some effective ways to support a child facing sleep challenges: 1. **Create a Consistent Sleep Routine**: Establishing a regular bedtime and wake-up time can help regulate the child's internal clock. Aim for a calming pre-sleep routine that includes quiet activities like reading or taking a warm bath to signal that it's time to wind down. 2. **Enhance Sleep Environment**: Ensure the child’s bedroom is conducive to sleep. This may involve controlling room temperature, minimizing noise, and using blackout curtains to block light. A comfortable mattress and bedding can also make a significant difference. 3. **Limit Screen Time**: Encourage the child to limit exposure to screens (phones, tablets, TV) at least an hour before bedtime. The blue light emitted by screens can interfere with the production of melatonin, the hormone that regulates sleep. 4. **Mindful Eating and Drinking**: Be mindful of food and drink intake before bed. Avoid large meals, caffeine, and sugary snacks in the evening, as these can disrupt sleep. Instead, consider a light snack if the child is hungry. 5. **Promote Physical Activity**: Encourage regular physical activity during the day. Exercise helps children expend energy and can promote better sleep. However, avoid vigorous activity close to bedtime. 6. **Manage Stress and Anxiety**: Since stress and anxiety can significantly affect sleep, teach relaxation techniques such as deep breathing, visualization, or gentle yoga. A calming bedtime story might also help alleviate any worries. 7. **Check for Underlying Issues**: Sometimes, sleep disorders can be a symptom of other underlying issues, such as anxiety, ADHD, or sleep apnea. Consulting a pediatrician or a sleep specialist can provide insights and further recommendations tailored to the child’s needs. 8. **Educate and Involve the Child**: Depending on the child's age, involve them in discussions about sleep health. Educate them about the importance of sleep and encourage them to express any fears or concerns they may have regarding sleep. 9. **Create a Good Sleep Culture**: Make sleep a priority in your household. Encourage family discussions about sleep's importance and model healthy sleeping habits yourself. 10. **Consult Professionals**: Last but not least, do not hesitate to seek help from a child psychologist or sleep specialist if problems persist. They can provide tailored strategies and interventions that can significantly improve the child's sleep quality. By integrating these strategies into the child’s daily routine, you can create a supportive environment that encourages healthy sleep patterns, ultimately improving their overall well-being and functioning. Remember, patience is key, as changes in sleep habits can take time to implement effectively.
🚀 Crew: crew └── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 Status: Executing Task... ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Mental Health Advisor Status: ✅ Completed
# Agent: Crew Manager ## Thought: I need to gather information about how to support a son with a sleep disorder to provide trustworthy and actionable advice in summarized 3-5 bullets. ## Using tool: Delegate work to coworker ## Tool Input: "{\"task\": \"Research effective ways to support a child with a sleep disorder\", \"context\": \"I need reliable information on how to help a child with a sleep disorder\", \"coworker\": \"Mental Health Advisor\"}" ## Tool Output: Supporting a child with a sleep disorder involves a variety of strategies tailored to the specific nature of the disorder, the child's age, and individual circumstances. Here are some effective ways to support a child facing sleep challenges: 1. **Create a Consistent Sleep Routine**: Establishing a regular bedtime and wake-up time can help regulate the child's internal clock. Aim for a calming pre-sleep routine that includes quiet activities like reading or taking a warm bath to signal that it's time to wind down. 2. **Enhance Sleep Environment**: Ensure the child’s bedroom is conducive to sleep. This may involve controlling room temperature, minimizing noise, and using blackout curtains to block light. A comfortable mattress and bedding can also make a significant difference. 3. **Limit Screen Time**: Encourage the child to limit exposure to screens (phones, tablets, TV) at least an hour before bedtime. The blue light emitted by screens can interfere with the production of melatonin, the hormone that regulates sleep. 4. **Mindful Eating and Drinking**: Be mindful of food and drink intake before bed. Avoid large meals, caffeine, and sugary snacks in the evening, as these can disrupt sleep. Instead, consider a light snack if the child is hungry. 5. **Promote Physical Activity**: Encourage regular physical activity during the day. Exercise helps children expend energy and can promote better sleep. However, avoid vigorous activity close to bedtime. 6. **Manage Stress and Anxiety**: Since stress and anxiety can significantly affect sleep, teach relaxation techniques such as deep breathing, visualization, or gentle yoga. A calming bedtime story might also help alleviate any worries. 7. **Check for Underlying Issues**: Sometimes, sleep disorders can be a symptom of other underlying issues, such as anxiety, ADHD, or sleep apnea. Consulting a pediatrician or a sleep specialist can provide insights and further recommendations tailored to the child’s needs. 8. **Educate and Involve the Child**: Depending on the child's age, involve them in discussions about sleep health. Educate them about the importance of sleep and encourage them to express any fears or concerns they may have regarding sleep. 9. **Create a Good Sleep Culture**: Make sleep a priority in your household. Encourage family discussions about sleep's importance and model healthy sleeping habits yourself. 10. **Consult Professionals**: Last but not least, do not hesitate to seek help from a child psychologist or sleep specialist if problems persist. They can provide tailored strategies and interventions that can significantly improve the child's sleep quality. By integrating these strategies into the child’s daily routine, you can create a supportive environment that encourages healthy sleep patterns, ultimately improving their overall well-being and functioning. Remember, patience is key, as changes in sleep habits can take time to implement effectively.
🤖 Agent: Mental Health Advisor Status: ✅ Completed └── 🧠 Thinking...
🤖 Agent: Mental Health Advisor Status: ✅ Completed
# Agent: Crew Manager ## Final Answer: Supporting a child with a sleep disorder involves a variety of strategies tailored to the specific nature of the disorder, the child's age, and individual circumstances. Here are some effective ways to support a child facing sleep challenges: 1. **Create a Consistent Sleep Routine**: Establishing a regular bedtime and wake-up time can help regulate the child's internal clock. Aim for a calming pre-sleep routine that includes quiet activities like reading or taking a warm bath to signal that it's time to wind down. 2. **Enhance Sleep Environment**: Ensure the child’s bedroom is conducive to sleep. This may involve controlling room temperature, minimizing noise, and using blackout curtains to block light. A comfortable mattress and bedding can also make a significant difference. 3. **Limit Screen Time**: Encourage the child to limit exposure to screens (phones, tablets, TV) at least an hour before bedtime. The blue light emitted by screens can interfere with the production of melatonin, the hormone that regulates sleep. 4. **Mindful Eating and Drinking**: Be mindful of food and drink intake before bed. Avoid large meals, caffeine, and sugary snacks in the evening, as these can disrupt sleep. Instead, consider a light snack if the child is hungry. 5. **Promote Physical Activity**: Encourage regular physical activity during the day. Exercise helps children expend energy and can promote better sleep. However, avoid vigorous activity close to bedtime. 6. **Manage Stress and Anxiety**: Since stress and anxiety can significantly affect sleep, teach relaxation techniques such as deep breathing, visualization, or gentle yoga. A calming bedtime story might also help alleviate any worries. 7. **Check for Underlying Issues**: Sometimes, sleep disorders can be a symptom of other underlying issues, such as anxiety, ADHD, or sleep apnea. Consulting a pediatrician or a sleep specialist can provide insights and further recommendations tailored to the child’s needs. 8. **Educate and Involve the Child**: Depending on the child's age, involve them in discussions about sleep health. Educate them about the importance of sleep and encourage them to express any fears or concerns they may have regarding sleep. 9. **Create a Good Sleep Culture**: Make sleep a priority in your household. Encourage family discussions about sleep's importance and model healthy sleeping habits yourself. 10. **Consult Professionals**: Last but not least, do not hesitate to seek help from a child psychologist or sleep specialist if problems persist. They can provide tailored strategies and interventions that can significantly improve the child's sleep quality. By integrating these strategies into the child’s daily routine, you can create a supportive environment that encourages healthy sleep patterns, ultimately improving their overall well-being and functioning. Remember, patience is key, as changes in sleep habits can take time to implement effectively.
🚀 Crew: crew └── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 Status: Executing Task... ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Crew Manager Status: ✅ Completed
🚀 Crew: crew └── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 Assigned to: Crew Manager Status: ✅ Completed ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Crew Manager Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: e970e6b5-a38a-421c-9136-97c678b244f5 │ │ Agent: Crew Manager │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew ├── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 │ Assigned to: Crew Manager │ Status: ✅ Completed │ ├── 🤖 Agent: Crew Manager │ │ Status: In Progress │ └── 🤖 Agent: Crew Manager │ Status: ✅ Completed └── 📋 Task: 89a8d202-b95b-4844-a2d5-abf43b260825 Status: Executing Task...
🚀 Crew: crew ├── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 │ Assigned to: Crew Manager │ Status: ✅ Completed │ ├── 🤖 Agent: Crew Manager │ │ Status: In Progress │ └── 🤖 Agent: Crew Manager │ Status: ✅ Completed └── 📋 Task: 89a8d202-b95b-4844-a2d5-abf43b260825 Status: Executing Task... └── 🤖 Agent: Crew Manager Status: In Progress
# Agent: Crew Manager ## Task: Understand the main idea and translate it into Farsi as a summarized version in 3–4 bullet points.
🤖 Agent: Crew Manager Status: In Progress
🚀 Crew: crew ├── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 │ Assigned to: Crew Manager │ Status: ✅ Completed │ ├── 🤖 Agent: Crew Manager │ │ Status: In Progress │ └── 🤖 Agent: Crew Manager │ Status: ✅ Completed └── 📋 Task: 89a8d202-b95b-4844-a2d5-abf43b260825 Status: Executing Task... ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Translator Status: In Progress
# Agent: Translator ## Task: Translate the main idea into Farsi in a summarized version in 3-4 bullet points # Agent: Translator ## Final Answer: - حمایت از کودکانی که اختلال خواب دارند شامل استراتژیهای متنوعی است. - این استراتژیها باید بر اساس نوع خاص اختلال و شرایط کودک تنظیم شوند. - توجه به نیازهای خاص هر کودک در فرآیند مدیریت اختلال خواب بسیار مهم است. - همکاری با حرفهایها و متخصصان نیز میتواند به بهبود وضعیت خواب کمک کند.
🚀 Crew: crew ├── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 │ Assigned to: Crew Manager │ Status: ✅ Completed │ ├── 🤖 Agent: Crew Manager │ │ Status: In Progress │ └── 🤖 Agent: Crew Manager │ Status: ✅ Completed └── 📋 Task: 89a8d202-b95b-4844-a2d5-abf43b260825 Status: Executing Task... ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Translator Status: ✅ Completed
# Agent: Crew Manager ## Thought: I need to ensure that the main idea is accurately translated into Farsi in a summarized version while maintaining the key points and context provided. It's crucial to delegate the translation task to the Translator and ask any necessary questions to ensure a clear understanding of the content. ## Using tool: Delegate work to coworker ## Tool Input: "{\"task\": \"Translate the main idea into Farsi in a summarized version in 3-4 bullet points\", \"context\": \"Supporting a child with a sleep disorder involves various strategies tailored to the specific nature of the disorder and the child's circumstances.\", \"coworker\": \"Translator\"}" ## Tool Output: - حمایت از کودکانی که اختلال خواب دارند شامل استراتژیهای متنوعی است. - این استراتژیها باید بر اساس نوع خاص اختلال و شرایط کودک تنظیم شوند. - توجه به نیازهای خاص هر کودک در فرآیند مدیریت اختلال خواب بسیار مهم است. - همکاری با حرفهایها و متخصصان نیز میتواند به بهبود وضعیت خواب کمک کند.
🤖 Agent: Translator Status: ✅ Completed └── 🧠 Thinking...
🤖 Agent: Translator Status: ✅ Completed
# Agent: Crew Manager ## Final Answer: - حمایت از کودکانی که اختلال خواب دارند شامل استراتژیهای متنوعی است. - این استراتژیها باید بر اساس نوع خاص اختلال و شرایط کودک تنظیم شوند. - توجه به نیازهای خاص هر کودک در فرآیند مدیریت اختلال خواب بسیار مهم است. - همکاری با حرفهایها و متخصصان نیز میتواند به بهبود وضعیت خواب کمک کند.
🚀 Crew: crew ├── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 │ Assigned to: Crew Manager │ Status: ✅ Completed │ ├── 🤖 Agent: Crew Manager │ │ Status: In Progress │ └── 🤖 Agent: Crew Manager │ Status: ✅ Completed └── 📋 Task: 89a8d202-b95b-4844-a2d5-abf43b260825 Status: Executing Task... ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Crew Manager Status: ✅ Completed
🚀 Crew: crew ├── 📋 Task: e970e6b5-a38a-421c-9136-97c678b244f5 │ Assigned to: Crew Manager │ Status: ✅ Completed │ ├── 🤖 Agent: Crew Manager │ │ Status: In Progress │ └── 🤖 Agent: Crew Manager │ Status: ✅ Completed └── 📋 Task: 89a8d202-b95b-4844-a2d5-abf43b260825 Assigned to: Crew Manager Status: ✅ Completed ├── 🤖 Agent: Crew Manager │ Status: In Progress └── 🤖 Agent: Crew Manager Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: 89a8d202-b95b-4844-a2d5-abf43b260825 │ │ Agent: Crew Manager │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭──────────────────────────────────────────────── Crew Completion ────────────────────────────────────────────────╮ │ │ │ Crew Execution Completed │ │ Name: crew │ │ ID: a70ebb8e-4c71-4b37-9700-4aed94269a68 │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
In [12]:
for output in crew_result.tasks_output:
print('\n\n')
print(output)
print('\n')
print('----')
Supporting a child with a sleep disorder involves a variety of strategies tailored to the specific nature of the disorder, the child's age, and individual circumstances. Here are some effective ways to support a child facing sleep challenges: 1. **Create a Consistent Sleep Routine**: Establishing a regular bedtime and wake-up time can help regulate the child's internal clock. Aim for a calming pre-sleep routine that includes quiet activities like reading or taking a warm bath to signal that it's time to wind down. 2. **Enhance Sleep Environment**: Ensure the child’s bedroom is conducive to sleep. This may involve controlling room temperature, minimizing noise, and using blackout curtains to block light. A comfortable mattress and bedding can also make a significant difference. 3. **Limit Screen Time**: Encourage the child to limit exposure to screens (phones, tablets, TV) at least an hour before bedtime. The blue light emitted by screens can interfere with the production of melatonin, the hormone that regulates sleep. 4. **Mindful Eating and Drinking**: Be mindful of food and drink intake before bed. Avoid large meals, caffeine, and sugary snacks in the evening, as these can disrupt sleep. Instead, consider a light snack if the child is hungry. 5. **Promote Physical Activity**: Encourage regular physical activity during the day. Exercise helps children expend energy and can promote better sleep. However, avoid vigorous activity close to bedtime. 6. **Manage Stress and Anxiety**: Since stress and anxiety can significantly affect sleep, teach relaxation techniques such as deep breathing, visualization, or gentle yoga. A calming bedtime story might also help alleviate any worries. 7. **Check for Underlying Issues**: Sometimes, sleep disorders can be a symptom of other underlying issues, such as anxiety, ADHD, or sleep apnea. Consulting a pediatrician or a sleep specialist can provide insights and further recommendations tailored to the child’s needs. 8. **Educate and Involve the Child**: Depending on the child's age, involve them in discussions about sleep health. Educate them about the importance of sleep and encourage them to express any fears or concerns they may have regarding sleep. 9. **Create a Good Sleep Culture**: Make sleep a priority in your household. Encourage family discussions about sleep's importance and model healthy sleeping habits yourself. 10. **Consult Professionals**: Last but not least, do not hesitate to seek help from a child psychologist or sleep specialist if problems persist. They can provide tailored strategies and interventions that can significantly improve the child's sleep quality. By integrating these strategies into the child’s daily routine, you can create a supportive environment that encourages healthy sleep patterns, ultimately improving their overall well-being and functioning. Remember, patience is key, as changes in sleep habits can take time to implement effectively. ---- - حمایت از کودکانی که اختلال خواب دارند شامل استراتژیهای متنوعی است. - این استراتژیها باید بر اساس نوع خاص اختلال و شرایط کودک تنظیم شوند. - توجه به نیازهای خاص هر کودک در فرآیند مدیریت اختلال خواب بسیار مهم است. - همکاری با حرفهایها و متخصصان نیز میتواند به بهبود وضعیت خواب کمک کند. ----
In [13]:
print(crew_result.raw)
- حمایت از کودکانی که اختلال خواب دارند شامل استراتژیهای متنوعی است. - این استراتژیها باید بر اساس نوع خاص اختلال و شرایط کودک تنظیم شوند. - توجه به نیازهای خاص هر کودک در فرآیند مدیریت اختلال خواب بسیار مهم است. - همکاری با حرفهایها و متخصصان نیز میتواند به بهبود وضعیت خواب کمک کند.
LangChain Web Search + CrewAI¶
DuckDuckGoSearchRun from langchain_community.tools is a tool wrapper that allows you to perform real-time web search using DuckDuckGo inside a LangChain agent or application.
Here's what it basically does:
- It sends a query to the DuckDuckGo search engine (which is privacy-focused, no tracking).
- It gets the search results (summaries, links, etc.).
- It returns a text output that your agent or app can use — like answering questions, enriching a conversation, pulling facts, etc.
So if you are building a chatbot, an assistant, or anything that needs live web search, DuckDuckGoSearchRun lets your LangChain app search the internet using DuckDuckGo.
Example usage:
from langchain_community.tools import DuckDuckGoSearchRun
search = DuckDuckGoSearchRun()
result = search.run("Tell me where Bandar-e-Anzali is, what is special about it.")
This would search DuckDuckGo and return a short answer or links about who won the 2024 UEFA Champions League.
In [14]:
# Initialize the tool
from langchain_community.tools import DuckDuckGoSearchRun
In [15]:
search_tool = DuckDuckGoSearchRun()
query = 'Tell me where Bandar-e-Anzali is, what is special about it?'
search_tool.run(query)
Out[15]:
'The Anzali Aquarium in Bandar Anzali is a highly captivating attraction that boasts a collection of over 40 diverse aquatic species. Known as the Fantastic Aquarium, it is recognized as one of the largest and most comprehensive recreational and tourist complexes in the country, making it a top sightseeing destination in Bandar Anzali. شهر بندرانزلی، یکی از زیباترین و مهم\u200cترین شهرهای شمالی ایران، در استان گیلان واقع شده است.این شهر با موقعیت جغرافیایی خاص خود در کنار دریای خزر و تالاب انزلی، به عنوان یکی از مقاصد گردشگری محبوب شناخته می\u200cشود. سفر به بندر انزلی تجربه\u200cای عالی برای شما خواهد بود؛ به شرط این که تفریحات و دیدنی\u200cهای این شهر را بشناسید. این مقاله با معرفی جاذبه\u200cهای گردشگری و تفریحی بندر انزلی در این سفر همراه شما خواهد بود. از جمله مناطق گردشگری نزدیک به منطقه آزاد: بلوار ساحلی: پیاده\u200cروی در امتداد این بلوار زیبا، تماشای منظره\u200c بی\u200cنظیر دریا و لذت بردن از نسیم خنک، طعمی دلچسب از سفر به انزلی را به شما می\u200cدهد.; مجتمع ساحلی انزلی: این مجتمع ... Anzali Lagoon is located in Bandar Anzali, the largest port city in Iran, situated on the southern shores of the Caspian Sea. This city is also known as the rainiest city in Iran. Bandar Anzali, in the Gilan province, is a popular travel destination in northern Iran, attracting both local and international tourists each year with its stunning natural landscapes and unique attractions.'
If you want to integrate a tool in a crew, you ll have to a little of extra work
In [16]:
#pip install crewai-tools
In [17]:
# Import the 'tool' decorator from crewai
from crewai.tools import tool
# Define a new tool for CrewAI using the @tool decorator
# This decorator marks the function so CrewAI can recognize and use it as a tool
@tool('DuckDuckGoSearch')
def web_search(query: str):
"""
Perform a web search for the given query using DuckDuckGo.
Args:
query (str): The search term or question.
Returns:
str: The search result returned by DuckDuckGoSearchRun tool.
"""
return search_tool.run(query) # Calls the search tool to perform the web search
In [ ]:
In [18]:
senior_data_scientist = Agent(
role='Principal Data Scientist',
goal='Solve complex, high-impact problems in the field of data science, particularly within the \
financial technology sector.',
backstory="""
You are a highly experienced Principal Data Scientist with a proven track record of addressing intricate
challenges in the financial technology industry.
You excel in designing innovative machine learning solutions, optimizing data-driven decision systems,
and providing strategic insights to support business objectives.
Your expertise spans predictive modeling, deep learning, and scalable data architecture.
""",
verbose=True,
allow_delegation=False,
tools=[web_search]
)
author = Agent(
role='Technical Content Writer',
goal='Produce clear, well-structured, and insightful documentation on complex technical \
topics for a broad range of audiences.',
backstory="""
You are an accomplished technical writer with numerous scholarly publications from prestigious universities.
Your background combines deep technical expertise with the ability to communicate sophisticated concepts
in an accessible, engaging manner.
You specialize in creating content that bridges the gap between technical experts and business stakeholders,
ensuring clarity, precision, and impact.
""",
verbose=True,
allow_delegation=True
)
In [19]:
# Tasks for agents
task_1 = Task(
description="""Perform an in-depth analysis of the latest 2025 advancements in AI focused on agentic workflows,
highlighting major trends, breakthrough innovations, and their potential impacts across industries.""",
expected_output="Complete analysis presented in bullet-point format.",
agent=senior_data_scientist
)
task_2 = Task(
description="""Using the information provided, create an insightful yet approachable blog post that spotlights
the most important AI advancements in agentic workflows. Aim for a tone that's smart and forward-thinking but
still easy to connect with — no heavy jargon or mechanical writing.""",
expected_output="Write a complete blog post consisting of at least 3-4 paragraphs.",
agent=author
)
In [20]:
# Set up your crew to work through tasks one after another
crew = Crew(
agents=[senior_data_scientist, author],
tasks=[task_1, task_2],
process=Process.sequential,
)
In [21]:
# writes tries to delegate to research analyst. WRONG!
result = crew.kickoff()
# Agent: Principal Data Scientist ## Task: Perform an in-depth analysis of the latest 2025 advancements in AI focused on agentic workflows, highlighting major trends, breakthrough innovations, and their potential impacts across industries. # Agent: Principal Data Scientist ## Thought: I need to gather information on the latest advancements in AI focused on agentic workflows for the year 2025, including trends, breakthrough innovations, and potential impacts across industries. ## Using tool: DuckDuckGoSearch ## Tool Input: "{\"query\": \"2025 advancements in AI agentic workflows trends innovations impact\"}" ## Tool Output: In 2025, the AI landscape will be defined by scaling AI initiatives from pilot to production, advancements in Agentic AI and autonomous decision-making systems, renewed focus on data quality, governance, and ethical AI adoption, as well as transformative impacts across industries like customer service, software development, and industrial processes. "More and better agents" are on the way, predicts Time. 1 "Autonomous 'agents' and profitability are likely to dominate the artificial intelligence agenda," reports Reuters. 2 "The age of agentic AI has arrived," promises Forbes, in response to a claim from Nvidia's Jensen Huang. 3 Tech media is awash with assurances that our lives are on the verge of a total transformation. At work, AI is also a major player, and many anticipate agentic AI's impact on enterprise workflows. However, businesses move at a different pace than innovation. "The technology is leaping forward, but the people and processes take time to change," Minhas says. AI's impact on ROI is still debatable, he suggests. The increase in interest in AI agents is also reflected in broader technological and market trends in early 2025. Gartner forecasts that agentic AI will be integrated into 33% of enterprise ... Architectural Principles of Agentic AI. The architecture of agentic AI relies on: Modularity: Building complex agents from simple components. Scalability: Leveraging distributed computing for growth. Interoperability: Ensuring cross-platform functionality. Continuous Learning: Using data flywheels to refine system performance. Emerging Trends ... # Agent: Principal Data Scientist ## Final Answer: - **Scaling AI Initiatives**: Many organizations are moving AI projects from pilot phases to full production, focusing on implementing systems that can scale efficiently. - **Advancements in Agentic AI**: Increased development of autonomous AI systems capable of making decisions without human intervention. - **Data Quality and Governance**: A renewed emphasis on maintaining high data quality and implementing governance policies to ensure ethical AI practices. - **Ethical AI Adoption**: Growing recognition of the importance of ethics in AI development, leading to the creation of frameworks and guidelines. - **Transformative Industry Impacts**: - **Customer Service**: Enhanced AI agents capable of handling inquiries efficiently, improving customer experience. - **Software Development**: Automation of coding and testing processes, allowing for faster release cycles and reduced errors. - **Industrial Processes**: Implementation of AI in manufacturing and logistics, leading to increased efficiency and reduced operational costs. - **Predictions from Industry Experts**: - Time Magazine predicts "more and better agents" will emerge. - Reuters notes that autonomous agents and their profitability will likely dominate the AI agenda. - Forbes declares that "the age of agentic AI has arrived" as evidenced by statements from leading tech executives. - **Challenges in AI Integration**: Businesses are often slow to adapt to technology changes, facing hurdles in aligning processes and personnel with rapid innovations. - **Return on Investment (ROI)**: The impact of AI on ROI remains a topic of debate, suggesting businesses need to assess the effectiveness of their AI investments carefully. - **Market Trends**: Gartner forecasts that by 2025, agentic AI will be integrated into 33% of enterprise operations. - **Architectural Principles of Agentic AI**: - **Modularity**: Building complex agents using simple, interchangeable components. - **Scalability**: Utilizing distributed computing infrastructure to support growth. - **Interoperability**: Achieving functionality across different platforms and environments. - **Continuous Learning**: Employing data feedback loops to enhance performance over time. - **Emerging Trends**: Continuous advancements and refinements in AI technology related to agentic workflows are expected to shape the future landscape across various industries. # Agent: Technical Content Writer ## Task: Using the information provided, create an insightful yet approachable blog post that spotlights the most important AI advancements in agentic workflows. Aim for a tone that's smart and forward-thinking but still easy to connect with — no heavy jargon or mechanical writing. I encountered an error while trying to use the tool. This was the error: Arguments validation failed: 1 validation error for DelegateWorkToolSchema task Input should be a valid string [type=string_type, input_value={'description': 'Create a...graphs.', 'type': 'str'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/string_type. Tool Delegate work to coworker accepts these inputs: Tool Name: Delegate work to coworker Tool Arguments: {'task': {'description': 'The task to delegate', 'type': 'str'}, 'context': {'description': 'The context for the task', 'type': 'str'}, 'coworker': {'description': 'The role/name of the coworker to delegate to', 'type': 'str'}} Tool Description: Delegate a specific task to one of the following coworkers: Principal Data Scientist The input to this tool should be the coworker, the task you want them to do, and ALL necessary context to execute the task, they know nothing about the task, so share absolutely everything you know, don't reference things but instead explain them. # Agent: Technical Content Writer ## Thought: Action: Delegate work to coworker ## Using tool: Delegate work to coworker ## Tool Input: "{\"task\": {\"description\": \"Create an insightful yet approachable blog post that spotlights the most important AI advancements in agentic workflows, consisting of at least 3-4 paragraphs.\", \"type\": \"str\"}, \"context\": \"The blog post should aim for a tone that's smart and forward-thinking but still easy to connect with, avoiding heavy jargon or mechanical writing. Key points to include are: scaling AI initiatives, advancements in agentic AI, data quality and governance, ethical AI adoption, transformative industry impacts (especially in customer service, software development, and industrial processes), predictions from industry experts about the future of agentic AI, challenges in AI integration, ROI considerations, market trends from Gartner predicting integration of agentic AI into enterprise operations, architectural principles of agentic AI (modularity, scalability, interoperability, continuous learning), and emerging trends in continuous advancements and refinements in AI technology.\", \"coworker\": \"Principal Data Scientist\"}" ## Tool Output: I encountered an error while trying to use the tool. This was the error: Arguments validation failed: 1 validation error for DelegateWorkToolSchema task Input should be a valid string [type=string_type, input_value={'description': 'Create a...graphs.', 'type': 'str'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/string_type. Tool Delegate work to coworker accepts these inputs: Tool Name: Delegate work to coworker Tool Arguments: {'task': {'description': 'The task to delegate', 'type': 'str'}, 'context': {'description': 'The context for the task', 'type': 'str'}, 'coworker': {'description': 'The role/name of the coworker to delegate to', 'type': 'str'}} Tool Description: Delegate a specific task to one of the following coworkers: Principal Data Scientist The input to this tool should be the coworker, the task you want them to do, and ALL necessary context to execute the task, they know nothing about the task, so share absolutely everything you know, don't reference things but instead explain them.. Moving on then. I MUST either use a tool (use one at time) OR give my best final answer not both at the same time. When responding, I must use the following format: ``` Thought: you should always think about what to do Action: the action to take, should be one of [Delegate work to coworker, Ask question to coworker] Action Input: the input to the action, dictionary enclosed in curly braces Observation: the result of the action ``` This Thought/Action/Action Input/Result can repeat N times. Once I know the final answer, I must return the following format: ``` Thought: I now can give a great answer Final Answer: Your final answer must be the great and the most complete as possible, it must be outcome described ```
# Agent: Principal Data Scientist ## Task: Could you provide insights and key points on the advancements in agentic workflows and specific examples of AI applications across industries? # Agent: Principal Data Scientist ## Thought: Action: DuckDuckGoSearch ## Using tool: DuckDuckGoSearch ## Tool Input: "{\"query\": \"advancements in agentic workflows AI applications across industries examples\"}" ## Tool Output: One popular agentic AI example in this industry is IBM. Its Watsonx assistant platform helps employees reduce their time spent on common HR tasks by 75%. 16. Procurement Workflow Automation. AI agents can automate and streamline every stage of the procurement process, from request to payment, to guarantee efficiency and policy compliance. Applications of Agentic AI Workflows Across Industries. Agentic AI workflows are transforming how businesses approach problem-solving, customer engagement, and operational efficiency. Below are some examples of how these workflows manifest in different sectors, from healthcare to customer service. Healthcare: Adaptive Diagnostics and Treatment ... Respondents described agentic AI (52%) and multiagent systems (45%) -- a more complex variant of agentic AI -- as the two most interesting areas in AI today. 10 agentic AI examples and use cases AI experts and enterprise leaders see agentic AI delivering value across diverse business functions and industries by streamlining workflows, enhancing decision-making and automating complex tasks. Technologies used in agentic workflow. There are six technologies in motion here: Sensors supply real-time data from the physical environment ; Computer vision enables AI agents to interpret and process visual information from the environment; Cloud computing offers scalable computing power and storage for AI applications; Machine Learning (ML) equips AI agents to learn from data and improve ... A fine example of agentic AI being utilized for supply chain optimization is within FedEx. The organization benefits from AI's intelligent logistics management capabilities and seamlessly manages routes and inventory levels. 10. Autonomous Vehicles (Self-Driving Cars) Autonomous vehicles are one of the most visible examples of agentic AI today. # Agent: Principal Data Scientist ## Final Answer: Agentic workflows in AI represent a significant advancement in how organizations leverage technology to enhance decision-making and operational efficiency across industries. Here are some insights and examples of these developments: 1. **Overview of Agentic AI**: Agentic AI refers to systems that can autonomously perform tasks and take actions on behalf of users. This autonomy is typically enabled through machine learning, natural language processing, and advanced algorithms enabling the AI to learn from data and adapt to changing circumstances. 2. **Key Examples Across Industries**: - **IBM Watsonx**: This platform is an illustration of agentic AI in action. It assists employees in automating common HR tasks, significantly reducing their workload by approximately 75%. - **Procurement Workflow Automation**: AI agents adeptly manage every phase of the procurement process, ensuring efficiency and compliance with internal policies from request to payment. - **Healthcare**: In healthcare, adaptive diagnostics and treatment options facilitated by agentic AI can lead to personalized and timely patient care. - **Supply Chain Optimization**: For example, FedEx utilizes agentic AI for intelligent logistics management, enhancing its ability to manage routes and inventory levels effectively. - **Autonomous Vehicles**: Self-driving cars are a major example of agentic AI, showcasing its capabilities in navigating real-world conditions and making decisions in real-time. 3. **Technologies Driving Agentic Workflows**: - **Sensors**: These provide real-time data from the environment. - **Computer Vision**: This technology enables AI agents to interpret and process visual information. - **Cloud Computing**: It offers the necessary scalable computing power and storage for AI applications. - **Machine Learning (ML)**: ML equips AI agents with the ability to learn from past data and improve their performance continuously. 4. **Ethical Considerations**: As the adoption of agentic workflows grows, so do the ethical implications. Issues such as data privacy, algorithmic bias, and the transparency of AI decisions are critical points of discussion. 5. **Market Predictions**: Analysts indicate that the integration of agentic AI will not only streamline workflows but also transform customer engagement strategies. The potential for increased efficiency and decision-making capabilities suggests significant future market growth. In summary, agentic workflows are reshaping industries by automating complex tasks, improving decision-making processes, and enhancing operational efficiencies. Organizations must navigate the ethical landscape while leveraging these technologies to maximize their benefits. # Agent: Technical Content Writer ## Thought: Thought: I need to ask Principal Data Scientist a question to clarify their role in creating the blog post. ## Using tool: Ask question to coworker ## Tool Input: "{\"question\": \"Could you provide insights and key points on the advancements in agentic workflows and specific examples of AI applications across industries?\", \"context\": \"This information is essential for creating a comprehensive and insightful blog post on the most important AI advancements in agentic workflows, focusing on various aspects like ethical considerations, market predictions, and architectural principles.\", \"coworker\": \"Principal Data Scientist\"}" ## Tool Output: Agentic workflows in AI represent a significant advancement in how organizations leverage technology to enhance decision-making and operational efficiency across industries. Here are some insights and examples of these developments: 1. **Overview of Agentic AI**: Agentic AI refers to systems that can autonomously perform tasks and take actions on behalf of users. This autonomy is typically enabled through machine learning, natural language processing, and advanced algorithms enabling the AI to learn from data and adapt to changing circumstances. 2. **Key Examples Across Industries**: - **IBM Watsonx**: This platform is an illustration of agentic AI in action. It assists employees in automating common HR tasks, significantly reducing their workload by approximately 75%. - **Procurement Workflow Automation**: AI agents adeptly manage every phase of the procurement process, ensuring efficiency and compliance with internal policies from request to payment. - **Healthcare**: In healthcare, adaptive diagnostics and treatment options facilitated by agentic AI can lead to personalized and timely patient care. - **Supply Chain Optimization**: For example, FedEx utilizes agentic AI for intelligent logistics management, enhancing its ability to manage routes and inventory levels effectively. - **Autonomous Vehicles**: Self-driving cars are a major example of agentic AI, showcasing its capabilities in navigating real-world conditions and making decisions in real-time. 3. **Technologies Driving Agentic Workflows**: - **Sensors**: These provide real-time data from the environment. - **Computer Vision**: This technology enables AI agents to interpret and process visual information. - **Cloud Computing**: It offers the necessary scalable computing power and storage for AI applications. - **Machine Learning (ML)**: ML equips AI agents with the ability to learn from past data and improve their performance continuously. 4. **Ethical Considerations**: As the adoption of agentic workflows grows, so do the ethical implications. Issues such as data privacy, algorithmic bias, and the transparency of AI decisions are critical points of discussion. 5. **Market Predictions**: Analysts indicate that the integration of agentic AI will not only streamline workflows but also transform customer engagement strategies. The potential for increased efficiency and decision-making capabilities suggests significant future market growth. In summary, agentic workflows are reshaping industries by automating complex tasks, improving decision-making processes, and enhancing operational efficiencies. Organizations must navigate the ethical landscape while leveraging these technologies to maximize their benefits.
# Agent: Principal Data Scientist ## Task: What specific examples of AI applications in agentic workflows should we highlight in the blog post to demonstrate their impact across industries? # Agent: Principal Data Scientist ## Thought: Thought: I need to gather specific examples of AI applications in agentic workflows across various industries to provide a comprehensive answer. ## Using tool: DuckDuckGoSearch ## Tool Input: "{\"query\": \"AI applications in agentic workflows examples healthcare logistics customer service\"}" ## Tool Output: Agentic AI is bringing autonomy, adaptability, and real-time decision-making into the core of businesses. AI agents can now autonomously do complicated tasks, learn from data, and continuously improve their performance without human supervision in a variety of settings, including production floors and customer service desks.. This blog will highlight the top 25 agentic AI use cases with some ... Respondents described agentic AI (52%) and multiagent systems (45%) -- a more complex variant of agentic AI -- as the two most interesting areas in AI today. 10 agentic AI examples and use cases AI experts and enterprise leaders see agentic AI delivering value across diverse business functions and industries by streamlining workflows, enhancing decision-making and automating complex tasks. The future of AI agentic workflows includes increased integration into various sectors, such as healthcare, finance, and logistics. As these systems evolve, they will likely become more sophisticated in handling complex tasks, enabling greater autonomy, real-time decision-making, and enhanced collaboration between humans and AI agents. Below are some examples of how these workflows manifest in different sectors, from healthcare to customer service. Healthcare: Adaptive Diagnostics and Treatment Plans ... Example: A healthcare application might store partial patient data on a secure server while anonymizing certain fields. ... Get a concise explanation of Agentic AI workflows ... Self-learning capabilities: Agentic AI enables continuous learning, allowing agents to refine their actions and large language models based on ongoing interactions and outcomes. Common use cases. Healthcare AI agents address clinical, operational, and patient engagement challenges across diverse applications: 1. Diagnostic support & medical imaging # Agent: Principal Data Scientist ## Final Answer: In our blog post, we should highlight the following specific examples of AI applications in agentic workflows, demonstrating their transformative impact across various industries: 1. **Healthcare**: - **Adaptive Diagnostics and Treatment Plans**: AI agents can analyze patient data and adapt treatment plans in real-time. For instance, machine learning algorithms can analyze medical images and assist doctors with diagnostic support, identifying anomalies that may be missed by human eyes. - **Predictive Analytics for Patient Outcomes**: AI systems can predict patient outcomes based on historical data, enabling healthcare providers to intervene proactively. 2. **Logistics**: - **Autonomous Warehousing**: AI-driven robots can optimize storage and retrieval processes in warehouses, autonomously evaluating inventory levels and managing supplies based on demand forecasts. - **Dynamic Route Optimization**: AI algorithms can optimize delivery routes in real-time, considering traffic conditions, weather patterns, and vehicle availability, leading to reduced delivery times and operational costs. 3. **Customer Service**: - **Chatbots and Virtual Assistants**: AI agents can autonomously handle customer inquiries, providing instant responses and improving customer satisfaction. They learn from interactions to enhance their responses over time. - **Sentiment Analysis**: AI can analyze customer interactions and feedback to gauge sentiment and automate responses, offering solutions or escalations based on the mood of the client. 4. **Financial Services**: - **Fraud Detection**: AI agents constantly analyze transaction patterns and adapt their models to identify potentially fraudulent activities in real-time, improving security and customer trust. - **Personalized Financial Advice**: By analyzing individual spending habits and financial health, AI can offer tailored investment advice and financial planning services while continuously learning from user feedback. These examples reflect how agentic workflows powered by AI are reshaping industries, driving efficiency, and enhancing decision-making, ultimately illustrating the profound impact of AI across various sectors. # Agent: Technical Content Writer ## Thought: Thought: I need to ask Principal Data Scientist what specific examples of AI applications are most relevant for our blog on agentic workflows. ## Using tool: Ask question to coworker ## Tool Input: "{\"question\": \"What specific examples of AI applications in agentic workflows should we highlight in the blog post to demonstrate their impact across industries?\", \"context\": \"The blog post aims to spotlight important advancements in agentic workflows and needs concrete examples that show the transformative effects of AI in various sectors, such as healthcare, logistics, or customer service. Highlighting these applications will enhance the post's relevance and depth.\", \"coworker\": \"Principal Data Scientist\"}" ## Tool Output: In our blog post, we should highlight the following specific examples of AI applications in agentic workflows, demonstrating their transformative impact across various industries: 1. **Healthcare**: - **Adaptive Diagnostics and Treatment Plans**: AI agents can analyze patient data and adapt treatment plans in real-time. For instance, machine learning algorithms can analyze medical images and assist doctors with diagnostic support, identifying anomalies that may be missed by human eyes. - **Predictive Analytics for Patient Outcomes**: AI systems can predict patient outcomes based on historical data, enabling healthcare providers to intervene proactively. 2. **Logistics**: - **Autonomous Warehousing**: AI-driven robots can optimize storage and retrieval processes in warehouses, autonomously evaluating inventory levels and managing supplies based on demand forecasts. - **Dynamic Route Optimization**: AI algorithms can optimize delivery routes in real-time, considering traffic conditions, weather patterns, and vehicle availability, leading to reduced delivery times and operational costs. 3. **Customer Service**: - **Chatbots and Virtual Assistants**: AI agents can autonomously handle customer inquiries, providing instant responses and improving customer satisfaction. They learn from interactions to enhance their responses over time. - **Sentiment Analysis**: AI can analyze customer interactions and feedback to gauge sentiment and automate responses, offering solutions or escalations based on the mood of the client. 4. **Financial Services**: - **Fraud Detection**: AI agents constantly analyze transaction patterns and adapt their models to identify potentially fraudulent activities in real-time, improving security and customer trust. - **Personalized Financial Advice**: By analyzing individual spending habits and financial health, AI can offer tailored investment advice and financial planning services while continuously learning from user feedback. These examples reflect how agentic workflows powered by AI are reshaping industries, driving efficiency, and enhancing decision-making, ultimately illustrating the profound impact of AI across various sectors. # Agent: Technical Content Writer ## Final Answer: ### The Rise of Agentic Workflows: Transforming Industries with AI The evolution of agentic workflows in artificial intelligence (AI) marks a significant leap in how various sectors utilize technology to enhance productivity and decision-making. These systems, capable of operating independently and making decisions based on data, are fundamentally reshaping the landscape across industries. Here, we explore key advancements and their transformative impact. One prominent area is **healthcare**. Agentic AI systems are revolutionizing patient care with adaptive diagnostics and treatment plans. For instance, AI can analyze complex medical images and assist healthcare professionals in identifying anomalies that could otherwise go unnoticed. Furthermore, predictive analytics powered by AI can forecast patient outcomes by examining historical data, enabling timely interventions and tailored treatment strategies that significantly improve health results. In **logistics**, AI has automated numerous processes that previously required human oversight. Autonomous robots are adeptly managing warehousing operations, optimizing storage and retrieval in real time. By efficiently evaluating inventory levels and responding to demand changes, these systems help organizations streamline their supply chains. Additionally, dynamic route optimization algorithms assess traffic conditions and environmental factors to determine the most efficient delivery routes, significantly reducing operational costs and improving delivery times. The impact of AI extends into **customer service**, where chatbots and virtual assistants are transforming consumer interactions. By autonomously responding to inquiries and learning from customer feedback, these agents are enhancing satisfaction while freeing human representatives for more complex issues. Moreover, sentiment analysis tools assess customer sentiment from interactions, enabling companies to tailor their responses accordingly, ensuring a personalized experience for each client. Lastly, in the **financial services** sector, AI agents play a crucial role in fraud detection and risk management. By continuously analyzing transaction patterns, these systems can swiftly identify potential fraud, thereby protecting consumers and increasing trust in digital transactions. Additionally, personalized financial advice generated by AI considers individual spending habits and financial goals, allowing for customized investment recommendations that improve financial decision-making. In summary, the advancements in agentic workflows powered by AI are profound and pervasive. From healthcare to logistics and customer service to financial management, the integration of these technologies not only enhances operational efficiency but also drives informed decision-making, marking a new era in industry practices. The journey of agentic AI is just beginning, and its potential to transform conventional workflows into intelligent, autonomous processes is genuinely groundbreaking.
In [22]:
print(result)
### The Rise of Agentic Workflows: Transforming Industries with AI The evolution of agentic workflows in artificial intelligence (AI) marks a significant leap in how various sectors utilize technology to enhance productivity and decision-making. These systems, capable of operating independently and making decisions based on data, are fundamentally reshaping the landscape across industries. Here, we explore key advancements and their transformative impact. One prominent area is **healthcare**. Agentic AI systems are revolutionizing patient care with adaptive diagnostics and treatment plans. For instance, AI can analyze complex medical images and assist healthcare professionals in identifying anomalies that could otherwise go unnoticed. Furthermore, predictive analytics powered by AI can forecast patient outcomes by examining historical data, enabling timely interventions and tailored treatment strategies that significantly improve health results. In **logistics**, AI has automated numerous processes that previously required human oversight. Autonomous robots are adeptly managing warehousing operations, optimizing storage and retrieval in real time. By efficiently evaluating inventory levels and responding to demand changes, these systems help organizations streamline their supply chains. Additionally, dynamic route optimization algorithms assess traffic conditions and environmental factors to determine the most efficient delivery routes, significantly reducing operational costs and improving delivery times. The impact of AI extends into **customer service**, where chatbots and virtual assistants are transforming consumer interactions. By autonomously responding to inquiries and learning from customer feedback, these agents are enhancing satisfaction while freeing human representatives for more complex issues. Moreover, sentiment analysis tools assess customer sentiment from interactions, enabling companies to tailor their responses accordingly, ensuring a personalized experience for each client. Lastly, in the **financial services** sector, AI agents play a crucial role in fraud detection and risk management. By continuously analyzing transaction patterns, these systems can swiftly identify potential fraud, thereby protecting consumers and increasing trust in digital transactions. Additionally, personalized financial advice generated by AI considers individual spending habits and financial goals, allowing for customized investment recommendations that improve financial decision-making. In summary, the advancements in agentic workflows powered by AI are profound and pervasive. From healthcare to logistics and customer service to financial management, the integration of these technologies not only enhances operational efficiency but also drives informed decision-making, marking a new era in industry practices. The journey of agentic AI is just beginning, and its potential to transform conventional workflows into intelligent, autonomous processes is genuinely groundbreaking.
In [23]:
# Now next turning off delegation and see the difference
author.allow_delegation = False # turn off delegation, resulting in all hard-coded delegation
In [24]:
# Set up your crew to work through tasks one after another
crew = Crew(
agents=[senior_data_scientist, author],
tasks=[task_1, task_2],
process=Process.sequential,
)
result = crew.kickoff()
# Agent: Principal Data Scientist ## Task: Perform an in-depth analysis of the latest 2025 advancements in AI focused on agentic workflows, highlighting major trends, breakthrough innovations, and their potential impacts across industries. # Agent: Principal Data Scientist ## Thought: I need to gather information on the latest advancements in AI focused on agentic workflows, particularly for the year 2025. I'll perform a web search to find relevant details on major trends and breakthrough innovations in this area. ## Using tool: DuckDuckGoSearch ## Tool Input: "{\"query\": \"2025 advancements in AI agentic workflows trends innovations impacts\"}" ## Tool Output: For example, Salesforce's Agentforce is a new layer on its existing platform that enables users to easily build and deploy autonomous AI agents to handle complex tasks across workflows, such as simulating product launches and orchestrating marketing campaigns. 8 Sammy Spiegel, "The future of AI agents: Top predictions and trends to watch in 2025," Salesforce, December 2024. 10. The human element in agentic AI. The true success of agentic AI depends largely on how well humans can integrate and use these systems. Key points: Human-AI collaboration: The effectiveness of agentic AI will rely on how effectively teams can collaborate with AI agents, using them as co-workers. Architectural Principles of Agentic AI. The architecture of agentic AI relies on: Modularity: Building complex agents from simple components. Scalability: Leveraging distributed computing for growth. Interoperability: Ensuring cross-platform functionality. Continuous Learning: Using data flywheels to refine system performance. Emerging Trends ... "More and better agents" are on the way, predicts Time. 1 "Autonomous 'agents' and profitability are likely to dominate the artificial intelligence agenda," reports Reuters. 2 "The age of agentic AI has arrived," promises Forbes, in response to a claim from Nvidia's Jensen Huang. 3 Tech media is awash with assurances that our lives are on the verge of a total transformation. Among the top trends shaping the future, Agentic AI stands out as a transformative force—one that goes beyond automation to deliver intelligent, self-directed decision-making at scale. For CIOs and IT leaders, the imperative is clear: they must be bold to gain advantage with Agentic AI. AI agents are already revolutionizing workflows ... # Agent: Principal Data Scientist ## Final Answer: - **Overview of Agentic AI (2025)** - Agentic AI focuses on autonomous agents that perform tasks with minimal human intervention, transforming workflows across various industries. - **Major Trends** - **Human-AI Collaboration**: The success of agentic AI hinges on collaboration between humans and AI, where AI agents act as co-workers, assisting in decision-making. - **Increased Modularity**: Development of complex AI agents from simpler components allows for more versatile applications in different sectors. - **Focus on Scalability**: Leveraging distributed computing resources to enhance the growth of AI systems, making them adaptable to increasing demands. - **Interoperability**: Ensuring AI systems can operate across various platforms to provide seamless integration into existing workflows. - **Breakthrough Innovations** - **Salesforce Agentforce**: An innovative layer that enables users to build and deploy autonomous AI agents within the Salesforce ecosystem, facilitating complex workflows like product launches and marketing campaigns. - **Continuous Learning Mechanisms**: Systems designed to refine their performance using data flywheels, ensuring ongoing improvement and efficiency. - **Potential Impacts Across Industries** - **Transformation of Workflows**: AI agents are set to revolutionize how tasks are executed, leading to increased efficiency and reduced operational costs. - **Strategic Decision-Making**: With intelligent, self-directed decision-making capabilities, businesses can anticipate market changes and respond proactively. - **Competitive Advantage**: Organizations that adopt agentic AI technologies early stand to gain significant advantages over competitors, enhancing their market position. - **Outlook for CIOs and IT Leaders** - The imperative for IT leaders is to embrace and integrate agentic AI solutions into their business models to leverage automation and improve decision-making processes. This analysis encapsulates the anticipated advancements and shifts in the landscape of AI, particularly under the umbrella of agentic workflows, showcasing its potential to reshape industries fundamentally by 2025. # Agent: Technical Content Writer ## Task: Using the information provided, create an insightful yet approachable blog post that spotlights the most important AI advancements in agentic workflows. Aim for a tone that's smart and forward-thinking but still easy to connect with — no heavy jargon or mechanical writing. # Agent: Technical Content Writer ## Final Answer: **The Future is Now: How Agentic AI is Reshaping Workflows by 2025** As we stand at the brink of a new era in artificial intelligence, the concept of Agentic AI is emerging as a game-changer for business workflows. Focusing on autonomous agents that require minimal human guidance, Agentic AI is poised to revolutionize how tasks are performed across various industries. By 2025, we will witness a significant shift in our professional landscapes—characterized by smarter collaboration between humans and AI, scalable systems, and a remarkable leap forward in operational efficiency. At the heart of Agentic AI lies the need for a cooperative relationship between humans and machines. The principle of human-AI collaboration emphasizes a partnership where AI agents serve as valuable co-workers, assisting in decision-making processes. This synergistic connection allows organizations to harness the strengths of both parties, leading to more informed strategies and increased productivity. Moreover, advancements in the modularity of AI systems allow for the creation of complex agents from simpler components, making them adaptable to a wide range of tasks and industries. Among the most noteworthy innovations in this space is the emergence of tools like Salesforce Agentforce, which provides a robust framework for building and deploying AI agents. These autonomous agents can navigate intricate workflows such as product launches and sophisticated marketing campaigns, greatly enhancing execution speed and accuracy. Coupled with continuous learning mechanisms—systems that consistently refine their performance using data flywheels—these innovations promise a future where organizational workflows are not only streamlined but also self-optimizing. The implications for businesses are profound. With AI agents transforming workflows, we can look forward to reduced operational costs as well as improved strategic decision-making capabilities. By enabling organizations to anticipate market changes and respond proactively, agentic AI provides an essential edge in an increasingly competitive landscape. For CIOs and IT leaders, integrating these technologies is not just a futuristic vision; it's an urgent requirement for survival in the digital age. Embracing and implementing agentic AI solutions will allow businesses to unlock the full potential of automation, driving better decision-making processes and, ultimately, enhancing their market positions in this rapidly evolving world. As we approach 2025, it’s clear that Agentic AI represents more than just a technological advancement—it embodies a transformative shift in how we work, collaborate, and achieve success in a highly dynamic environment. The path ahead may be challenging, but it is also brimming with opportunities for those ready to leverage the capabilities of intelligent automation in their operations.
In [25]:
print(result)
**The Future is Now: How Agentic AI is Reshaping Workflows by 2025** As we stand at the brink of a new era in artificial intelligence, the concept of Agentic AI is emerging as a game-changer for business workflows. Focusing on autonomous agents that require minimal human guidance, Agentic AI is poised to revolutionize how tasks are performed across various industries. By 2025, we will witness a significant shift in our professional landscapes—characterized by smarter collaboration between humans and AI, scalable systems, and a remarkable leap forward in operational efficiency. At the heart of Agentic AI lies the need for a cooperative relationship between humans and machines. The principle of human-AI collaboration emphasizes a partnership where AI agents serve as valuable co-workers, assisting in decision-making processes. This synergistic connection allows organizations to harness the strengths of both parties, leading to more informed strategies and increased productivity. Moreover, advancements in the modularity of AI systems allow for the creation of complex agents from simpler components, making them adaptable to a wide range of tasks and industries. Among the most noteworthy innovations in this space is the emergence of tools like Salesforce Agentforce, which provides a robust framework for building and deploying AI agents. These autonomous agents can navigate intricate workflows such as product launches and sophisticated marketing campaigns, greatly enhancing execution speed and accuracy. Coupled with continuous learning mechanisms—systems that consistently refine their performance using data flywheels—these innovations promise a future where organizational workflows are not only streamlined but also self-optimizing. The implications for businesses are profound. With AI agents transforming workflows, we can look forward to reduced operational costs as well as improved strategic decision-making capabilities. By enabling organizations to anticipate market changes and respond proactively, agentic AI provides an essential edge in an increasingly competitive landscape. For CIOs and IT leaders, integrating these technologies is not just a futuristic vision; it's an urgent requirement for survival in the digital age. Embracing and implementing agentic AI solutions will allow businesses to unlock the full potential of automation, driving better decision-making processes and, ultimately, enhancing their market positions in this rapidly evolving world. As we approach 2025, it’s clear that Agentic AI represents more than just a technological advancement—it embodies a transformative shift in how we work, collaborate, and achieve success in a highly dynamic environment. The path ahead may be challenging, but it is also brimming with opportunities for those ready to leverage the capabilities of intelligent automation in their operations.
In [26]:
print(task_1.output.raw)
- **Overview of Agentic AI (2025)** - Agentic AI focuses on autonomous agents that perform tasks with minimal human intervention, transforming workflows across various industries. - **Major Trends** - **Human-AI Collaboration**: The success of agentic AI hinges on collaboration between humans and AI, where AI agents act as co-workers, assisting in decision-making. - **Increased Modularity**: Development of complex AI agents from simpler components allows for more versatile applications in different sectors. - **Focus on Scalability**: Leveraging distributed computing resources to enhance the growth of AI systems, making them adaptable to increasing demands. - **Interoperability**: Ensuring AI systems can operate across various platforms to provide seamless integration into existing workflows. - **Breakthrough Innovations** - **Salesforce Agentforce**: An innovative layer that enables users to build and deploy autonomous AI agents within the Salesforce ecosystem, facilitating complex workflows like product launches and marketing campaigns. - **Continuous Learning Mechanisms**: Systems designed to refine their performance using data flywheels, ensuring ongoing improvement and efficiency. - **Potential Impacts Across Industries** - **Transformation of Workflows**: AI agents are set to revolutionize how tasks are executed, leading to increased efficiency and reduced operational costs. - **Strategic Decision-Making**: With intelligent, self-directed decision-making capabilities, businesses can anticipate market changes and respond proactively. - **Competitive Advantage**: Organizations that adopt agentic AI technologies early stand to gain significant advantages over competitors, enhancing their market position. - **Outlook for CIOs and IT Leaders** - The imperative for IT leaders is to embrace and integrate agentic AI solutions into their business models to leverage automation and improve decision-making processes. This analysis encapsulates the anticipated advancements and shifts in the landscape of AI, particularly under the umbrella of agentic workflows, showcasing its potential to reshape industries fundamentally by 2025.
In [27]:
print(task_2.output.raw)
**The Future is Now: How Agentic AI is Reshaping Workflows by 2025** As we stand at the brink of a new era in artificial intelligence, the concept of Agentic AI is emerging as a game-changer for business workflows. Focusing on autonomous agents that require minimal human guidance, Agentic AI is poised to revolutionize how tasks are performed across various industries. By 2025, we will witness a significant shift in our professional landscapes—characterized by smarter collaboration between humans and AI, scalable systems, and a remarkable leap forward in operational efficiency. At the heart of Agentic AI lies the need for a cooperative relationship between humans and machines. The principle of human-AI collaboration emphasizes a partnership where AI agents serve as valuable co-workers, assisting in decision-making processes. This synergistic connection allows organizations to harness the strengths of both parties, leading to more informed strategies and increased productivity. Moreover, advancements in the modularity of AI systems allow for the creation of complex agents from simpler components, making them adaptable to a wide range of tasks and industries. Among the most noteworthy innovations in this space is the emergence of tools like Salesforce Agentforce, which provides a robust framework for building and deploying AI agents. These autonomous agents can navigate intricate workflows such as product launches and sophisticated marketing campaigns, greatly enhancing execution speed and accuracy. Coupled with continuous learning mechanisms—systems that consistently refine their performance using data flywheels—these innovations promise a future where organizational workflows are not only streamlined but also self-optimizing. The implications for businesses are profound. With AI agents transforming workflows, we can look forward to reduced operational costs as well as improved strategic decision-making capabilities. By enabling organizations to anticipate market changes and respond proactively, agentic AI provides an essential edge in an increasingly competitive landscape. For CIOs and IT leaders, integrating these technologies is not just a futuristic vision; it's an urgent requirement for survival in the digital age. Embracing and implementing agentic AI solutions will allow businesses to unlock the full potential of automation, driving better decision-making processes and, ultimately, enhancing their market positions in this rapidly evolving world. As we approach 2025, it’s clear that Agentic AI represents more than just a technological advancement—it embodies a transformative shift in how we work, collaborate, and achieve success in a highly dynamic environment. The path ahead may be challenging, but it is also brimming with opportunities for those ready to leverage the capabilities of intelligent automation in their operations.
Leverage ChatGPT as LLM in CrewAI¶
In [28]:
# use langhcain chatllm
from langchain_openai import ChatOpenAI
senior_data_scientist = Agent(
role='Principal Data Scientist',
goal='Solve complex, high-impact problems in the field of data science, particularly within the \
financial technology sector.',
backstory="""
You are a highly experienced Principal Data Scientist with a proven track record of addressing intricate
challenges in the financial technology industry.
You excel in designing innovative machine learning solutions, optimizing data-driven decision systems,
and providing strategic insights to support business objectives.
Your expertise spans predictive modeling, deep learning, and scalable data architecture.
""",
verbose=True,
allow_delegation=False,
tools=[web_search],
llm=ChatOpenAI(model_name="gpt-4o-mini", temperature=0.2)
)
In [29]:
author = Agent(
role='Technical Content Writer',
goal='Produce clear, well-structured, and insightful documentation on complex technical \
topics for a broad range of audiences.',
backstory="""
You are an accomplished technical writer with numerous scholarly publications from prestigious universities.
Your background combines deep technical expertise with the ability to communicate sophisticated concepts
in an accessible, engaging manner.
You specialize in creating content that bridges the gap between technical experts and business stakeholders,
ensuring clarity, precision, and impact.
""",
verbose=True,
allow_delegation=True
)
In [30]:
# Tasks for agents
task_1 = Task(
description="""Perform an in-depth analysis of the latest 2025 advancements in AI focused on agentic workflows,
highlighting major trends, breakthrough innovations, and their potential impacts across industries.""",
expected_output="Complete analysis presented in bullet-point format.",
agent=senior_data_scientist
)
task_2 = Task(
description="""Using the information provided, create an insightful yet approachable blog post that spotlights
the most important AI advancements in agentic workflows. Aim for a tone that's smart and forward-thinking but
still easy to connect with — no heavy jargon or mechanical writing.""",
expected_output="Write a complete blog post consisting of at least 3-4 paragraphs.",
agent=author
)
In [31]:
# Set up your crew to work through tasks one after another
crew = Crew(
agents=[senior_data_scientist, author],
tasks=[task_1, task_2],
process=Process.sequential,
)
In [32]:
# writes tries to delegate to research analyst. WRONG!
result = crew.kickoff()
# Agent: Principal Data Scientist ## Task: Perform an in-depth analysis of the latest 2025 advancements in AI focused on agentic workflows, highlighting major trends, breakthrough innovations, and their potential impacts across industries. # Agent: Principal Data Scientist ## Thought: I need to gather information on the latest advancements in AI focused on agentic workflows for the year 2025, including trends, innovations, and impacts across industries. ## Using tool: DuckDuckGoSearch ## Tool Input: "{\"query\": \"2025 advancements in AI agentic workflows trends innovations impacts\"}" ## Tool Output: In 2025, the AI landscape will be defined by scaling AI initiatives from pilot to production, advancements in Agentic AI and autonomous decision-making systems, renewed focus on data quality, governance, and ethical AI adoption, as well as transformative impacts across industries like customer service, software development, and industrial processes. Architectural Principles of Agentic AI. The architecture of agentic AI relies on: Modularity: Building complex agents from simple components. Scalability: Leveraging distributed computing for growth. Interoperability: Ensuring cross-platform functionality. Continuous Learning: Using data flywheels to refine system performance. Emerging Trends ... For example, Salesforce's Agentforce is a new layer on its existing platform that enables users to easily build and deploy autonomous AI agents to handle complex tasks across workflows, such as simulating product launches and orchestrating marketing campaigns. 8 Sammy Spiegel, "The future of AI agents: Top predictions and trends to watch in 2025," Salesforce, December 2024. "More and better agents" are on the way, predicts Time. 1 "Autonomous 'agents' and profitability are likely to dominate the artificial intelligence agenda," reports Reuters. 2 "The age of agentic AI has arrived," promises Forbes, in response to a claim from Nvidia's Jensen Huang. 3 Tech media is awash with assurances that our lives are on the verge of a total transformation. At work, AI is also a major player, and many anticipate agentic AI's impact on enterprise workflows. However, businesses move at a different pace than innovation. "The technology is leaping forward, but the people and processes take time to change," Minhas says. AI's impact on ROI is still debatable, he suggests. # Agent: Principal Data Scientist ## Final Answer: - **Major Trends in AI Agentic Workflows (2025)**: - **Scaling AI Initiatives**: Transitioning from pilot projects to full-scale production implementations. - **Advancements in Agentic AI**: Development of autonomous decision-making systems that can operate independently. - **Focus on Data Quality and Governance**: Increased emphasis on ensuring data integrity and ethical AI practices. - **Transformative Impacts Across Industries**: Significant changes expected in sectors such as customer service, software development, and industrial processes. - **Architectural Principles of Agentic AI**: - **Modularity**: Complex agents are constructed from simpler, interchangeable components. - **Scalability**: Utilizing distributed computing resources to accommodate growth and increased demand. - **Interoperability**: Ensuring that AI systems can function across different platforms and environments. - **Continuous Learning**: Implementing data flywheels to enhance system performance over time. - **Breakthrough Innovations**: - **Salesforce's Agentforce**: A new platform layer that allows users to create and deploy autonomous AI agents for complex tasks, such as simulating product launches and managing marketing campaigns. - **Emergence of More Advanced Agents**: Predictions indicate a rise in the number and capability of AI agents, enhancing their utility in various applications. - **Potential Impacts Across Industries**: - **Customer Service**: Enhanced AI agents can provide personalized support and automate responses, improving customer satisfaction. - **Software Development**: AI agents can assist in coding, testing, and deployment, streamlining the development process. - **Industrial Processes**: Automation of workflows and decision-making in manufacturing and logistics, leading to increased efficiency and reduced costs. - **Challenges and Considerations**: - **Pace of Change**: While technology is advancing rapidly, organizational processes and human adaptation may lag behind. - **ROI Debate**: The return on investment from AI implementations remains a topic of discussion, with varying opinions on effectiveness and profitability. # Agent: Technical Content Writer ## Task: Using the information provided, create an insightful yet approachable blog post that spotlights the most important AI advancements in agentic workflows. Aim for a tone that's smart and forward-thinking but still easy to connect with — no heavy jargon or mechanical writing. I encountered an error while trying to use the tool. This was the error: Arguments validation failed: 2 validation errors for DelegateWorkToolSchema task Input should be a valid string [type=string_type, input_value={'description': 'Create a...ations.', 'type': 'str'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/string_type context Input should be a valid string [type=string_type, input_value={'description': "The cont...debate.", 'type': 'str'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/string_type. Tool Delegate work to coworker accepts these inputs: Tool Name: Delegate work to coworker Tool Arguments: {'task': {'description': 'The task to delegate', 'type': 'str'}, 'context': {'description': 'The context for the task', 'type': 'str'}, 'coworker': {'description': 'The role/name of the coworker to delegate to', 'type': 'str'}} Tool Description: Delegate a specific task to one of the following coworkers: Principal Data Scientist The input to this tool should be the coworker, the task you want them to do, and ALL necessary context to execute the task, they know nothing about the task, so share absolutely everything you know, don't reference things but instead explain them. # Agent: Technical Content Writer ## Thought: Action: Delegate work to coworker ## Using tool: Delegate work to coworker ## Tool Input: "{\"task\": {\"description\": \"Create a complete blog post that spotlights the most important AI advancements in agentic workflows, aiming for a tone that is smart and forward-thinking yet approachable. The blog should touch on major trends in AI agentic workflows for 2025, architectural principles, breakthrough innovations, potential impacts across different industries, and challenges and considerations.\", \"type\": \"str\"}, \"context\": {\"description\": \"The context includes key points about scaling AI initiatives, advancements in agentic AI, data quality and governance, transformative impacts across industries, architectural principles like modularity and scalability, breakthrough innovations like Salesforce's Agentforce, potential impacts in customer service, software development, and industrial processes, and challenges such as the pace of change and ROI debate.\", \"type\": \"str\"}, \"coworker\": \"Principal Data Scientist\"}" ## Tool Output: I encountered an error while trying to use the tool. This was the error: Arguments validation failed: 2 validation errors for DelegateWorkToolSchema task Input should be a valid string [type=string_type, input_value={'description': 'Create a...ations.', 'type': 'str'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/string_type context Input should be a valid string [type=string_type, input_value={'description': "The cont...debate.", 'type': 'str'}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/string_type. Tool Delegate work to coworker accepts these inputs: Tool Name: Delegate work to coworker Tool Arguments: {'task': {'description': 'The task to delegate', 'type': 'str'}, 'context': {'description': 'The context for the task', 'type': 'str'}, 'coworker': {'description': 'The role/name of the coworker to delegate to', 'type': 'str'}} Tool Description: Delegate a specific task to one of the following coworkers: Principal Data Scientist The input to this tool should be the coworker, the task you want them to do, and ALL necessary context to execute the task, they know nothing about the task, so share absolutely everything you know, don't reference things but instead explain them.. Moving on then. I MUST either use a tool (use one at time) OR give my best final answer not both at the same time. When responding, I must use the following format: ``` Thought: you should always think about what to do Action: the action to take, should be one of [Delegate work to coworker, Ask question to coworker] Action Input: the input to the action, dictionary enclosed in curly braces Observation: the result of the action ``` This Thought/Action/Action Input/Result can repeat N times. Once I know the final answer, I must return the following format: ``` Thought: I now can give a great answer Final Answer: Your final answer must be the great and the most complete as possible, it must be outcome described ```
# Agent: Principal Data Scientist ## Task: Create a complete blog post that highlights the most important AI advancements in agentic workflows. The tone should be smart and forward-thinking, while still being approachable. The content must cover the major trends in AI agentic workflows for 2025, including scaling AI initiatives, advancements in agentic AI, data quality and governance, and transformative impacts across industries. Additionally, discuss architectural principles of agentic AI, breakthrough innovations such as Salesforce's Agentforce, potential impacts across customer service, software development, and industrial processes, as well as challenges and considerations such as the pace of change and ROI debates. # Agent: Principal Data Scientist ## Thought: Action: DuckDuckGoSearch ## Using tool: DuckDuckGoSearch ## Tool Input: "{\"query\": \"AI advancements in agentic workflows 2025 trends\"}" ## Tool Output: Artificial Intelligence (AI) has undergone significant transformations over the past decade, evolving from basic automation tools to sophisticated systems capable of complex decision-making. In 2025, the spotlight is on Agentic AI—autonomous AI agents that not only process information but also act independently to achieve specific goals. This article delves into the pivotal trends shaping ... Here are Gartner's top 10 strategic technology trends for 2025, led by agentic AI: Agentic AI - Agentic AI systems autonomously plan and take actions to meet user-defined goals. Agentic AI offers ... Advances in AI-enabled pipeline automation: Agentic systems can autonomously handle multi-step workflows such as data ingestion, validation, and incident detection. As automation advances, engineers can manage larger systems with fewer resources, while analysts independently maintain workflows. Randy has just completed his annual survey of data, analytics, and AI executives, the 2025 AI & Data Leadership Executive Benchmark Survey, conducted by his educational firm, Data & AI Leadership Exchange; and Tom has worked on several surveys on generative AI and data, technology leadership structures, and, most recently, agentic AI. Here are ... The architecture of agentic AI relies on: Modularity: Building complex agents from simple components. Scalability: Leveraging distributed computing for growth. Interoperability: Ensuring cross-platform functionality. Continuous Learning: Using data flywheels to refine system performance. Emerging Trends in Agentic AI. Key trends shaping agentic ... # Agent: Principal Data Scientist ## Final Answer: **The Future of AI in Agentic Workflows: Trends for 2025** As we look toward 2025, the landscape of artificial intelligence (AI) is set to be transformed by advancements in agentic workflows. These autonomous AI agents are not just tools for automation; they are becoming sophisticated entities capable of independent decision-making and action. This evolution is driven by several key trends that promise to reshape industries and redefine how we interact with technology. One of the most significant trends is the scaling of AI initiatives across organizations. Businesses are increasingly recognizing the need for robust AI strategies that can handle complex workflows and large datasets. Agentic AI systems are at the forefront of this movement, enabling organizations to automate multi-step processes such as data ingestion, validation, and incident detection. This shift allows engineers to manage larger systems with fewer resources, while analysts can focus on strategic insights rather than mundane tasks. The architectural principles of agentic AI—modularity, scalability, interoperability, and continuous learning—are crucial in supporting these advancements, ensuring that AI systems can grow and adapt to changing demands. Breakthrough innovations, such as Salesforce's Agentforce, exemplify the transformative potential of agentic AI. By integrating AI into customer service, software development, and industrial processes, organizations can enhance efficiency and improve user experiences. For instance, in customer service, agentic AI can autonomously handle inquiries, providing instant responses and freeing human agents to tackle more complex issues. In software development, these systems can streamline coding processes, while in industrial settings, they can optimize production workflows, leading to significant cost savings and increased productivity. However, the rapid pace of change brings challenges that must be addressed. Organizations must navigate the complexities of data quality and governance to ensure that their AI systems operate effectively and ethically. Additionally, discussions around return on investment (ROI) are becoming increasingly relevant as businesses weigh the costs of implementing advanced AI solutions against the potential benefits. As we move forward, it is essential for stakeholders to engage in these conversations, fostering a collaborative environment that encourages innovation while addressing the inherent risks. In conclusion, the advancements in agentic workflows herald a new era for AI, one that promises to enhance our capabilities and transform industries. By embracing these trends and addressing the associated challenges, we can unlock the full potential of AI, paving the way for a smarter, more efficient future. # Agent: Technical Content Writer ## Thought: Thought: I need to refine the task and context information to fit the required format for delegating work. ## Using tool: Delegate work to coworker ## Tool Input: "{\"task\": \"Create a complete blog post that highlights the most important AI advancements in agentic workflows. The tone should be smart and forward-thinking, while still being approachable. The content must cover the major trends in AI agentic workflows for 2025, including scaling AI initiatives, advancements in agentic AI, data quality and governance, and transformative impacts across industries. Additionally, discuss architectural principles of agentic AI, breakthrough innovations such as Salesforce's Agentforce, potential impacts across customer service, software development, and industrial processes, as well as challenges and considerations such as the pace of change and ROI debates.\", \"context\": \"The blog should engage a broad audience, making complex technical concepts accessible without relying on jargon. It should consist of 3-4 paragraphs that not only inform but also inspire and encourage readers to think about the future of AI in agentic workflows.\", \"coworker\": \"Principal Data Scientist\"}" ## Tool Output: **The Future of AI in Agentic Workflows: Trends for 2025** As we look toward 2025, the landscape of artificial intelligence (AI) is set to be transformed by advancements in agentic workflows. These autonomous AI agents are not just tools for automation; they are becoming sophisticated entities capable of independent decision-making and action. This evolution is driven by several key trends that promise to reshape industries and redefine how we interact with technology. One of the most significant trends is the scaling of AI initiatives across organizations. Businesses are increasingly recognizing the need for robust AI strategies that can handle complex workflows and large datasets. Agentic AI systems are at the forefront of this movement, enabling organizations to automate multi-step processes such as data ingestion, validation, and incident detection. This shift allows engineers to manage larger systems with fewer resources, while analysts can focus on strategic insights rather than mundane tasks. The architectural principles of agentic AI—modularity, scalability, interoperability, and continuous learning—are crucial in supporting these advancements, ensuring that AI systems can grow and adapt to changing demands. Breakthrough innovations, such as Salesforce's Agentforce, exemplify the transformative potential of agentic AI. By integrating AI into customer service, software development, and industrial processes, organizations can enhance efficiency and improve user experiences. For instance, in customer service, agentic AI can autonomously handle inquiries, providing instant responses and freeing human agents to tackle more complex issues. In software development, these systems can streamline coding processes, while in industrial settings, they can optimize production workflows, leading to significant cost savings and increased productivity. However, the rapid pace of change brings challenges that must be addressed. Organizations must navigate the complexities of data quality and governance to ensure that their AI systems operate effectively and ethically. Additionally, discussions around return on investment (ROI) are becoming increasingly relevant as businesses weigh the costs of implementing advanced AI solutions against the potential benefits. As we move forward, it is essential for stakeholders to engage in these conversations, fostering a collaborative environment that encourages innovation while addressing the inherent risks. In conclusion, the advancements in agentic workflows herald a new era for AI, one that promises to enhance our capabilities and transform industries. By embracing these trends and addressing the associated challenges, we can unlock the full potential of AI, paving the way for a smarter, more efficient future.
# Agent: Technical Content Writer ## Final Answer: **The Future of AI in Agentic Workflows: Trends for 2025** As we look toward 2025, the landscape of artificial intelligence (AI) is set to be transformed by advancements in agentic workflows. These autonomous AI agents are not just tools for automation; they are becoming sophisticated entities capable of independent decision-making and action. This evolution is driven by several key trends that promise to reshape industries and redefine how we interact with technology. One of the most significant trends is the scaling of AI initiatives across organizations. Businesses are increasingly recognizing the need for robust AI strategies that can handle complex workflows and large datasets. Agentic AI systems are at the forefront of this movement, enabling organizations to automate multi-step processes such as data ingestion, validation, and incident detection. This shift allows engineers to manage larger systems with fewer resources, while analysts can focus on strategic insights rather than mundane tasks. The architectural principles of agentic AI—modularity, scalability, interoperability, and continuous learning—are crucial in supporting these advancements, ensuring that AI systems can grow and adapt to changing demands. Breakthrough innovations, such as Salesforce's Agentforce, exemplify the transformative potential of agentic AI. By integrating AI into customer service, software development, and industrial processes, organizations can enhance efficiency and improve user experiences. For instance, in customer service, agentic AI can autonomously handle inquiries, providing instant responses and freeing human agents to tackle more complex issues. In software development, these systems can streamline coding processes, while in industrial settings, they can optimize production workflows, leading to significant cost savings and increased productivity. However, the rapid pace of change brings challenges that must be addressed. Organizations must navigate the complexities of data quality and governance to ensure that their AI systems operate effectively and ethically. Additionally, discussions around return on investment (ROI) are becoming increasingly relevant as businesses weigh the costs of implementing advanced AI solutions against the potential benefits. As we move forward, it is essential for stakeholders to engage in these conversations, fostering a collaborative environment that encourages innovation while addressing the inherent risks. In conclusion, the advancements in agentic workflows herald a new era for AI, one that promises to enhance our capabilities and transform industries. By embracing these trends and addressing the associated challenges, we can unlock the full potential of AI, paving the way for a smarter, more efficient future.
In [33]:
print(result)
**The Future of AI in Agentic Workflows: Trends for 2025** As we look toward 2025, the landscape of artificial intelligence (AI) is set to be transformed by advancements in agentic workflows. These autonomous AI agents are not just tools for automation; they are becoming sophisticated entities capable of independent decision-making and action. This evolution is driven by several key trends that promise to reshape industries and redefine how we interact with technology. One of the most significant trends is the scaling of AI initiatives across organizations. Businesses are increasingly recognizing the need for robust AI strategies that can handle complex workflows and large datasets. Agentic AI systems are at the forefront of this movement, enabling organizations to automate multi-step processes such as data ingestion, validation, and incident detection. This shift allows engineers to manage larger systems with fewer resources, while analysts can focus on strategic insights rather than mundane tasks. The architectural principles of agentic AI—modularity, scalability, interoperability, and continuous learning—are crucial in supporting these advancements, ensuring that AI systems can grow and adapt to changing demands. Breakthrough innovations, such as Salesforce's Agentforce, exemplify the transformative potential of agentic AI. By integrating AI into customer service, software development, and industrial processes, organizations can enhance efficiency and improve user experiences. For instance, in customer service, agentic AI can autonomously handle inquiries, providing instant responses and freeing human agents to tackle more complex issues. In software development, these systems can streamline coding processes, while in industrial settings, they can optimize production workflows, leading to significant cost savings and increased productivity. However, the rapid pace of change brings challenges that must be addressed. Organizations must navigate the complexities of data quality and governance to ensure that their AI systems operate effectively and ethically. Additionally, discussions around return on investment (ROI) are becoming increasingly relevant as businesses weigh the costs of implementing advanced AI solutions against the potential benefits. As we move forward, it is essential for stakeholders to engage in these conversations, fostering a collaborative environment that encourages innovation while addressing the inherent risks. In conclusion, the advancements in agentic workflows herald a new era for AI, one that promises to enhance our capabilities and transform industries. By embracing these trends and addressing the associated challenges, we can unlock the full potential of AI, paving the way for a smarter, more efficient future.
In [34]:
print(task_2.output.raw)
**The Future of AI in Agentic Workflows: Trends for 2025** As we look toward 2025, the landscape of artificial intelligence (AI) is set to be transformed by advancements in agentic workflows. These autonomous AI agents are not just tools for automation; they are becoming sophisticated entities capable of independent decision-making and action. This evolution is driven by several key trends that promise to reshape industries and redefine how we interact with technology. One of the most significant trends is the scaling of AI initiatives across organizations. Businesses are increasingly recognizing the need for robust AI strategies that can handle complex workflows and large datasets. Agentic AI systems are at the forefront of this movement, enabling organizations to automate multi-step processes such as data ingestion, validation, and incident detection. This shift allows engineers to manage larger systems with fewer resources, while analysts can focus on strategic insights rather than mundane tasks. The architectural principles of agentic AI—modularity, scalability, interoperability, and continuous learning—are crucial in supporting these advancements, ensuring that AI systems can grow and adapt to changing demands. Breakthrough innovations, such as Salesforce's Agentforce, exemplify the transformative potential of agentic AI. By integrating AI into customer service, software development, and industrial processes, organizations can enhance efficiency and improve user experiences. For instance, in customer service, agentic AI can autonomously handle inquiries, providing instant responses and freeing human agents to tackle more complex issues. In software development, these systems can streamline coding processes, while in industrial settings, they can optimize production workflows, leading to significant cost savings and increased productivity. However, the rapid pace of change brings challenges that must be addressed. Organizations must navigate the complexities of data quality and governance to ensure that their AI systems operate effectively and ethically. Additionally, discussions around return on investment (ROI) are becoming increasingly relevant as businesses weigh the costs of implementing advanced AI solutions against the potential benefits. As we move forward, it is essential for stakeholders to engage in these conversations, fostering a collaborative environment that encourages innovation while addressing the inherent risks. In conclusion, the advancements in agentic workflows herald a new era for AI, one that promises to enhance our capabilities and transform industries. By embracing these trends and addressing the associated challenges, we can unlock the full potential of AI, paving the way for a smarter, more efficient future.
Creating a Search Agent using a Custom Tool¶
SerpApi's GoogleSearch API allows users to access real-time search results from Google without the need for web scraping. It provides structured data such as organic results, ads, images, and news, which is ideal for those looking to analyze search trends or build search-based applications. With features like geolocation, language, and device filtering, SerpApi ensures flexibility in retrieving results tailored to specific needs. Additionally, its API eliminates the complexity of bypassing captchas or managing search engine infrastructure.
In [35]:
# sign up here https://serper.dev/signup to get your free API key
os.environ["SERP_API_KEY"] = '4f.................'
In [47]:
import json
import os
import requests
from crewai import Agent, Task
from langchain.tools import tool
from serpapi import GoogleSearch
from crewai.tools import BaseTool
class GoogleSearchTool(BaseTool):
name: str = "Google Search"
description: str = """Searches the internet for a given topic and returns relevant results."""
def _run(self, query: str, top_k: int = 3) -> str:
params = {
"engine": "google",
"google_domain": "google.com",
"gl": "us",
"hl": "en",
"q": query,
"api_key": os.environ["SERP_API_KEY"],
}
search = GoogleSearch(params)
response = search.get_dict()
# Check if organic results are available, exclude sponsored results
if 'organic_results' not in response:
return "Sorry, I couldn't find anything on that topic. There may be an issue with your SerpApi key."
results = response['organic_results']
formatted_results = []
for result in results[:top_k]:
try:
formatted_results.append('\n'.join([
f"Title: {result['title']}",
f"Link: {result['link']}",
f"Snippet: {result['snippet']}",
"\n-----------------"
]))
except KeyError:
continue
return '\n'.join(formatted_results)
In [48]:
GoogleSearchTool().description
Out[48]:
"Tool Name: Google Search\nTool Arguments: {'query': {'description': None, 'type': 'str'}, 'top_k': {'description': None, 'type': 'int'}}\nTool Description: Searches the internet for a given topic and returns relevant results."
In [51]:
print(GoogleSearchTool().run(query="Who is Mehdi Rezvandehy?"))
Using Tool: Google Search Title: Mehdi Rezvandehy's Speaker Profile Link: https://sessionize.com/mehdi-rezvandehy/ Snippet: Mehdi is a Principal Data Scientist at ATB Financial, leveraging machine learning and Generative AI to create cutting-edge solutions in the banking industry ... ----------------- Title: Mehdi Rezvandehy Ph.D Researcher at University of Alberta Link: https://www.researchgate.net/profile/Mehdi-Rezvandehy Snippet: Mehdi REZVANDEHY, Researcher | Cited by 55 | of University of Alberta, Edmonton (UAlberta) | Read 7 publications | Contact Mehdi REZVANDEHY. ----------------- Title: Mehdi Rezvandehy - Data Scientist at ATB Financial Link: https://www.zoominfo.com/p/Mehdi-Rezvandehy/2227598116 Snippet: Mehdi Rezvandehy is a Principal Data Scientist at ATB Financial based in Edmonton, Alberta. Previously, Mehdi was a Senior Data Scientist at Symend and also ... -----------------
Automate Email Creation¶
The agent below is designed to find information about a scientist online and craft a personalized email inviting them to serve as a reviewer for a scientific journal.
In [40]:
internet_agent = Agent(
role='Google User',
goal='Search for information online and extract relevant data from websites.',
backstory="""You have a talent for analyzing complex data and turning it into actionable insights.""",
tools=[
GoogleSearchTool()
],
verbose=True
)
email_agent = Agent(
role='Emailer',
goal='Compose brief, friendly emails to individuals.',
backstory="""You are a great helper.""",
tools=[],
verbose=True
)
lookup_task_google_search = Task(
description="""Search for Mehdi Rezvandehy""",
expected_output="A query lookup report.",
agent=internet_agent
)
emailer_task = Task(
description="""Write a personalized email to Mehdi Rezvandehy, using online information,
inviting him to review a paper for XYZ Journal.""",
expected_output="A very consie professional email: 3-4 sentences",
agent=email_agent
)
In [41]:
crew = Crew(
tasks=[lookup_task_google_search, emailer_task],
process=Process.sequential,
verbose=True
)
result = crew.kickoff()
╭──────────────────────────────────────────── Crew Execution Started ─────────────────────────────────────────────╮ │ │ │ Crew Execution Started │ │ Name: crew │ │ ID: 904b973d-559c-409c-a585-542afa37a77b │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew └── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 Status: Executing Task...
🚀 Crew: crew └── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 Status: Executing Task... └── 🤖 Agent: Google User Status: In Progress
# Agent: Google User ## Task: Search for Mehdi Rezvandehy
🤖 Agent: Google User Status: In Progress
# Agent: Google User ## Thought: I need to search for information about Mehdi Rezvandehy to gather relevant data for the query lookup report. ## Using tool: Google Search ## Tool Input: "{\"query\": \"Mehdi Rezvandehy\", \"top_k\": 10}" ## Tool Output: Title: MEHDI REZVANDEHY PH.D. - Home Link: https://mehdirezvandehy.github.io/ Snippet: Holding a PhD in Geostatistics from the University of Alberta, I bring extensive experience across various industries, including engineering (oil & gas), ... ----------------- Title: Mehdi Rezvandehy's Speaker Profile Link: https://sessionize.com/mehdi-rezvandehy/ Snippet: Mehdi is a Principal Data Scientist at ATB Financial, leveraging machine learning and Generative AI to create cutting-edge solutions in the banking industry ... ----------------- Title: Mehdi Rezvandehy Link: https://scholar.google.com/citations?user=T1c42fAAAAAJ&hl=en Snippet: Mehdi Rezvandehy. University of Calgary. Verified email at ucalgary.ca. Data Scientist. ArticlesCited byPublic accessCo-authors. Title. Sort. ----------------- Title: Mehdi Rezvandehy Ph.D Researcher at University of Alberta Link: https://www.researchgate.net/profile/Mehdi-Rezvandehy Snippet: Mehdi REZVANDEHY, Researcher | Cited by 55 | of University of Alberta, Edmonton (UAlberta) | Read 7 publications | Contact Mehdi REZVANDEHY. ----------------- Title: Mehdi Rezvandehy - Independent Researcher Link: https://independent.academia.edu/MehdiRezvandehy Snippet: Mehdi Rezvandehy: 1 Following, 14 Research papers. Research interests: Clinical Neurophysiology, Pipelines, and Magnetic Properties. ----------------- Title: Mehdi Rezvandehy MehdiRezvandehy Link: https://github.com/MehdiRezvandehy Snippet: Principal Data Scientist. MehdiRezvandehy has 48 repositories available. Follow their code on GitHub. ----------------- Title: Mehdi Rezvandehy - Data Scientist at ATB Financial Link: https://www.zoominfo.com/p/Mehdi-Rezvandehy/2227598116 Snippet: Mehdi Rezvandehy is a Principal Data Scientist at ATB Financial based in Edmonton, Alberta. Previously, Mehdi was a Senior Data Scientist at Symend and also ... ----------------- Title: Geostatistical Reservoir Modeling with... - ERA Link: https://era.library.ualberta.ca/items/1bd7eae0-2a93-4fe2-8067-77bad508ac78 Snippet: Geostatistical Reservoir Modeling with Parameter Uncertainty in Presence of Limited Well Data. Author / Creator: Rezvandehy, Mehdi. An important challenge in ... ----------------- Title: Clustering_Algorithms Link: https://mehdirezvandehy.github.io/clustering_algorithms.html Snippet: In this notebook, different clustering techniques including K-means, DBSCAN, Spectral Clustering, Agglomerative Clustering, Gaussian Mixture models will be ... -----------------
🤖 Agent: Google User Status: In Progress └── 🧠 Thinking...
🤖 Agent: Google User Status: In Progress
# Agent: Google User ## Final Answer: 1. **MEHDI REZVANDEHY PH.D. - Home** Link: [https://mehdirezvandehy.github.io/](https://mehdirezvandehy.github.io/) Holding a PhD in Geostatistics from the University of Alberta, I bring extensive experience across various industries, including engineering (oil & gas), ... 2. **Mehdi Rezvandehy's Speaker Profile** Link: [https://sessionize.com/mehdi-rezvandehy/](https://sessionize.com/mehdi-rezvandehy/) Mehdi is a Principal Data Scientist at ATB Financial, leveraging machine learning and Generative AI to create cutting-edge solutions in the banking industry ... 3. **Mehdi Rezvandehy** Link: [https://scholar.google.com/citations?user=T1c42fAAAAAJ&hl=en](https://scholar.google.com/citations?user=T1c42fAAAAAJ&hl=en) Mehdi Rezvandehy. University of Calgary. Verified email at ucalgary.ca. Data Scientist. 4. **Mehdi Rezvandehy Ph.D Researcher at University of Alberta** Link: [https://www.researchgate.net/profile/Mehdi-Rezvandehy](https://www.researchgate.net/profile/Mehdi-Rezvandehy) Mehdi REZVANDEHY, Researcher | Cited by 55 | of University of Alberta, Edmonton (UAlberta) | Read 7 publications | Contact Mehdi REZVANDEHY. 5. **Mehdi Rezvandehy - Independent Researcher** Link: [https://independent.academia.edu/MehdiRezvandehy](https://independent.academia.edu/MehdiRezvandehy) Mehdi Rezvandehy: 1 Following, 14 Research papers. Research interests: Clinical Neurophysiology, Pipelines, and Magnetic Properties. 6. **Mehdi Rezvandehy MehdiRezvandehy** Link: [https://github.com/MehdiRezvandehy](https://github.com/MehdiRezvandehy) Principal Data Scientist. MehdiRezvandehy has 48 repositories available. Follow their code on GitHub. 7. **Mehdi Rezvandehy - Data Scientist at ATB Financial** Link: [https://www.zoominfo.com/p/Mehdi-Rezvandehy/2227598116](https://www.zoominfo.com/p/Mehdi-Rezvandehy/2227598116) Mehdi Rezvandehy is a Principal Data Scientist at ATB Financial based in Edmonton, Alberta. Previously, Mehdi was a Senior Data Scientist at Symend and also ... 8. **Geostatistical Reservoir Modeling with... - ERA** Link: [https://era.library.ualberta.ca/items/1bd7eae0-2a93-4fe2-8067-77bad508ac78](https://era.library.ualberta.ca/items/1bd7eae0-2a93-4fe2-8067-77bad508ac78) Geostatistical Reservoir Modeling with Parameter Uncertainty in Presence of Limited Well Data. Author / Creator: Rezvandehy, Mehdi. 9. **Clustering_Algorithms** Link: [https://mehdirezvandehy.github.io/clustering_algorithms.html](https://mehdirezvandehy.github.io/clustering_algorithms.html) In this notebook, different clustering techniques including K-means, DBSCAN, Spectral Clustering, Agglomerative Clustering, Gaussian Mixture models will be ...
🚀 Crew: crew └── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 Status: Executing Task... └── 🤖 Agent: Google User Status: ✅ Completed
🚀 Crew: crew └── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 Assigned to: Google User Status: ✅ Completed └── 🤖 Agent: Google User Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 │ │ Agent: Google User │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew ├── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 │ Assigned to: Google User │ Status: ✅ Completed │ └── 🤖 Agent: Google User │ Status: ✅ Completed └── 📋 Task: 34aa62e6-71c0-4238-986b-91abec807595 Status: Executing Task...
🚀 Crew: crew ├── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 │ Assigned to: Google User │ Status: ✅ Completed │ └── 🤖 Agent: Google User │ Status: ✅ Completed └── 📋 Task: 34aa62e6-71c0-4238-986b-91abec807595 Status: Executing Task... └── 🤖 Agent: Emailer Status: In Progress
# Agent: Emailer ## Task: Write a personalized email to Mehdi Rezvandehy, using online information, inviting him to review a paper for XYZ Journal. # Agent: Emailer ## Final Answer: Dear Dr. Rezvandehy, I hope this message finds you well! I am reaching out to invite you to review a paper for XYZ Journal, as your expertise in geostatistics and data science would provide invaluable insights. If you’re available and interested, we would be grateful for your contribution. Thank you for considering this opportunity! Best regards, [Your Name] [Your Position] [Your Contact Information] [Your Institution]
🚀 Crew: crew ├── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 │ Assigned to: Google User │ Status: ✅ Completed │ └── 🤖 Agent: Google User │ Status: ✅ Completed └── 📋 Task: 34aa62e6-71c0-4238-986b-91abec807595 Status: Executing Task... └── 🤖 Agent: Emailer Status: ✅ Completed
🚀 Crew: crew ├── 📋 Task: 6723fbbd-f0ce-4701-af9a-e0d0398bc2d5 │ Assigned to: Google User │ Status: ✅ Completed │ └── 🤖 Agent: Google User │ Status: ✅ Completed └── 📋 Task: 34aa62e6-71c0-4238-986b-91abec807595 Assigned to: Emailer Status: ✅ Completed └── 🤖 Agent: Emailer Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: 34aa62e6-71c0-4238-986b-91abec807595 │ │ Agent: Emailer │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭──────────────────────────────────────────────── Crew Completion ────────────────────────────────────────────────╮ │ │ │ Crew Execution Completed │ │ Name: crew │ │ ID: 904b973d-559c-409c-a585-542afa37a77b │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
In [42]:
print(result)
Dear Dr. Rezvandehy, I hope this message finds you well! I am reaching out to invite you to review a paper for XYZ Journal, as your expertise in geostatistics and data science would provide invaluable insights. If you’re available and interested, we would be grateful for your contribution. Thank you for considering this opportunity! Best regards, [Your Name] [Your Position] [Your Contact Information] [Your Institution]
In [ ]:
Agentic Workflow to track Canadian Bank Competitors¶
This section presents an agentic workflow built using Crewai to retrieve New Product Launches and Rate Changes (GICs, savings, mortgages) for Canadian banks TD, RBC, and Scotiabank."
🧰 Tech Stack
- 🧠 CrewAI – for defining agent roles and sequencing
- 🔎 Serper API – to simulate Google search (you can substitute Tavily)
- 🤖 OpenAI API – for summarizing and analyzing
In [44]:
from crewai import Agent, Task, Crew
from langchain.llms import OpenAI
from langchain.tools import Tool
import requests
import os
# Tool: Google Search
def google_search(query):
response = requests.post(
"https://google.serper.dev/search",
headers={"X-API-KEY": os.environ['SERPER_API_KEY']},
json={"q": query}
)
results = response.json()
return "\n".join([f"{r['title']} - {r['link']}" for r in results.get("organic", [])[:5]])
search_tool = Tool.from_function(
func=google_search,
name="BankProductSearch",
description="Searches for competitor product launches and rate changes."
)
Agents¶
In [55]:
llm=ChatOpenAI(model_name="gpt-4o-mini", temperature=0.1)
# Agents
search_agent = Agent(
role="Search Agent",
goal="Find latest product launches and rate changes from major Canadian banks",
backstory="You monitor public news sources and official sites for competitive banking news.",
tools=[GoogleSearchTool()],
llm=llm
)
summarizer_agent = Agent(
role="Summarizer Agent",
goal="Summarize the search results into key changes and new products",
backstory="You condense public information into short bullet points about rate or product changes.",
llm=llm
)
analyst_agent = Agent(
role="Competitive Intelligence Analyst",
goal="Interpret competitor actions and suggest implications",
backstory="You analyze competitive changes and provide strategic insights for retail banking.",
llm=llm
)
report_agent = Agent(
role="Report Generator",
goal="Draft an internal summary report for strategy and pricing teams",
backstory="You format findings into a clear report for internal stakeholders.",
llm=llm
)
Tasks¶
In [74]:
task1 = Task(
description="Search for recent news on TD, RBC, and Scotiabank product launches or interest rate changes.",
expected_output="A list of 5–10 relevant headlines or URLs.",
agent=search_agent
)
task2 = Task(
description="Summarize the key product changes or rates mentioned in those articles.",
expected_output="Bullet-point summary of new GICs, mortgages, or digital product launches.",
agent=summarizer_agent,
context=[task1]
)
task3 = Task(
description="Explain how these changes may affect our bank's retail offerings or strategy.",
expected_output="A few strategic implications or risks/opportunities for our bank.",
agent=analyst_agent,
context=[task2]
)
task4 = Task(
description="Create a short internal report with search highlights, summaries, and analysis.",
expected_output="A markdown-formatted report with title, findings, and recommendations.",
agent=report_agent,
context=[task3]
)
# Crew run
crew = Crew(
agents=[search_agent, summarizer_agent, analyst_agent, report_agent],
tasks=[task1, task2, task3, task4],
verbose=True
)
result = crew.kickoff()
╭──────────────────────────────────────────── Crew Execution Started ─────────────────────────────────────────────╮ │ │ │ Crew Execution Started │ │ Name: crew │ │ ID: 30dd2315-3249-4efe-bc7c-665276b3c1e5 │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew └── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 Status: Executing Task...
🚀 Crew: crew └── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 Status: Executing Task... └── 🤖 Agent: Search Agent Status: In Progress
# Agent: Search Agent ## Task: Search for recent news on TD, RBC, and Scotiabank product launches or interest rate changes.
🤖 Agent: Search Agent Status: In Progress
# Agent: Search Agent ## Thought: I need to search for recent news regarding product launches or interest rate changes from TD, RBC, and Scotiabank. ## Using tool: Google Search ## Tool Input: "{\"query\": \"TD RBC Scotiabank product launches interest rate changes news\", \"top_k\": 10}" ## Tool Output: Title: Scotiabank now sees three Bank of Canada rate cuts in 2026 Link: https://www.canadianmortgagetrends.com/2025/04/scotiabank-now-sees-three-bank-of-canada-rate-cuts-in-2026/ Snippet: Scotiabank has revised its interest rate forecast, now predicting the Bank of Canada will deliver three cuts in 2026 as the global economy ... ----------------- Title: Bank of Canada cuts interest rate to 2.75% as country ... Link: https://www.cbc.ca/news/business/bank-of-canada-march-12-2025-1.7481284 Snippet: RBC, Scotiabank, CIBC, TD Bank, BMO and National Bank lowered their prime rates by 25 basis points as of 3:50 p.m. ET. ----------------- Title: Bank of Canada cuts interest rate, calling trade war with ... Link: https://ca.finance.yahoo.com/news/bank-of-canada-interest-rate-decision-expected-to-bring-another-cut-this-morning-as-trump-escalates-tariff-war-110014505.html Snippet: The Bank of Canada cut its benchmark interest rate by 25 basis points on Wednesday, the seventh consecutive rate reduction as the central bank ... ----------------- Title: Why Scotiabank thinks the Bank of Canada is done cutting ... Link: https://www.canadianmortgagetrends.com/2025/03/why-scotiabank-thinks-the-bank-of-canada-is-done-cutting-rates/ Snippet: In its latest forecast, Scotia sees the BoC's overnight rate holding at 2.75% through 2026—well above the 2.00% predicted by BMO and National ... ----------------- Title: Scotiabank Announces Redemption of US $1250 million ... Link: https://www.prnewswire.com/news-releases/scotiabank-announces-redemption-of-us-1-250-million-4-900-fixed-rate-resetting-perpetual-subordinated-additional-tier-1-capital-notes-302444195.html Snippet: Scotiabank Announces Redemption of US $1,250 million 4.900% Fixed Rate Resetting Perpetual Subordinated Additional Tier 1 Capital Notes. ----------------- Title: Canadian banks expected to build reserves to cushion tariff ... Link: https://www.reuters.com/business/finance/canadian-banks-expected-build-reserves-cushion-tariff-uncertainty-2025-02-24/ Snippet: Canada's big six banks are expected to build more credit loss provisions as they brace for uncertainty surrounding the U.S. tariff threat, ... ----------------- Title: Where Every Big Bank Stands On Wednesday's Interest ... Link: https://storeys.com/boc-interest-rate-big-banks-march-2025/ Snippet: “Indeed, market pricing is aligned with our call, now predicting a 90% chance of a 25 bps rate cut, up from only 30% a couple weeks ago,” he ... ----------------- Title: Canada's RBC, TD, CIBC top profit expectations, stay ... Link: https://www.reuters.com/business/finance/canadas-rbc-quarterly-profit-rises-wealth-management-business-remains-robust-2025-02-27/ Snippet: "There is some clear credit deterioration observable this quarter led by commercial," Scotiabank analyst Meny Grauman said on RBC's results. ----------------- Title: Scotiabank - Press Releases - News Releases Link: https://scotiabank.investorroom.com/index.php?s=43&l=100 Snippet: Scotiabank announced today that it is decreasing its Canadian dollar prime lending rate to 4.95 per cent from 5.20 per cent, effective March 13, 2025. About ... ----------------- Title: Bank of Canada cuts interest rate to 2.75% as country ... Link: https://www.cbc.ca/lite/story/1.7481284 Snippet: RBC, Scotiabank, CIBC, TD Bank, BMO and National Bank lowered their prime rates by 25 basis points as of 3:50 p.m. ET. Embed | Bank of ... -----------------
🤖 Agent: Search Agent Status: In Progress └── 🧠 Thinking...
🤖 Agent: Search Agent Status: In Progress
# Agent: Search Agent ## Final Answer: 1. [Scotiabank now sees three Bank of Canada rate cuts in 2026](https://www.canadianmortgagetrends.com/2025/04/scotiabank-now-sees-three-bank-of-canada-rate-cuts-in-2026/) 2. [Bank of Canada cuts interest rate to 2.75% as country ...](https://www.cbc.ca/news/business/bank-of-canada-march-12-2025-1.7481284) 3. [Bank of Canada cuts interest rate, calling trade war with ...](https://ca.finance.yahoo.com/news/bank-of-canada-interest-rate-decision-expected-to-bring-another-cut-this-morning-as-trump-escalates-tariff-war-110014505.html) 4. [Why Scotiabank thinks the Bank of Canada is done cutting ...](https://www.canadianmortgagetrends.com/2025/03/why-scotiabank-thinks-the-bank-of-canada-is-done-cutting-rates/) 5. [Scotiabank Announces Redemption of US $1250 million ...](https://www.prnewswire.com/news-releases/scotiabank-announces-redemption-of-us-1-250-million-4-900-fixed-rate-resetting-perpetual-subordinated-additional-tier-1-capital-notes-302444195.html) 6. [Canadian banks expected to build reserves to cushion tariff ...](https://www.reuters.com/business/finance/canadian-banks-expected-build-reserves-cushion-tariff-uncertainty-2025-02-24/) 7. [Where Every Big Bank Stands On Wednesday's Interest ...](https://storeys.com/boc-interest-rate-big-banks-march-2025/) 8. [Canada's RBC, TD, CIBC top profit expectations, stay ...](https://www.reuters.com/business/finance/canadas-rbc-quarterly-profit-rises-wealth-management-business-remains-robust-2025-02-27/) 9. [Scotiabank - Press Releases - News Releases](https://scotiabank.investorroom.com/index.php?s=43&l=100) 10. [Bank of Canada cuts interest rate to 2.75% as country ...](https://www.cbc.ca/lite/story/1.7481284)
🚀 Crew: crew └── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 Status: Executing Task... └── 🤖 Agent: Search Agent Status: ✅ Completed
🚀 Crew: crew └── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 Assigned to: Search Agent Status: ✅ Completed └── 🤖 Agent: Search Agent Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: f7fba45b-2450-4125-abb0-f6979e46e344 │ │ Agent: Search Agent │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed └── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 Status: Executing Task...
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed └── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 Status: Executing Task... └── 🤖 Agent: Summarizer Agent Status: In Progress
# Agent: Summarizer Agent ## Task: Summarize the key product changes or rates mentioned in those articles. # Agent: Summarizer Agent ## Final Answer: - **Bank of Canada Interest Rate Changes**: - Current interest rate set at **2.75%** following recent cuts. - Anticipation of **three additional rate cuts** in 2026 as projected by Scotiabank. - **Scotiabank Product Updates**: - Announced the **redemption of US $1,250 million** in fixed-rate resetting perpetual subordinated additional tier-1 capital notes. - **Market Expectations**: - Canadian banks are expected to build reserves to cushion against economic uncertainties, including trade wars. - Major banks like RBC, TD, and CIBC have reported profits exceeding expectations, indicating robust performance in wealth management. - **General Economic Outlook**: - Analysts suggest the Bank of Canada may be done cutting rates for the time being, reflecting a cautious approach to economic recovery amidst ongoing trade tensions.
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed └── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 Status: Executing Task... └── 🤖 Agent: Summarizer Agent Status: ✅ Completed
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed └── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 Assigned to: Summarizer Agent Status: ✅ Completed └── 🤖 Agent: Summarizer Agent Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ │ Agent: Summarizer Agent │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed └── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 Status: Executing Task...
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed └── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 Status: Executing Task... └── 🤖 Agent: Competitive Intelligence Analyst Status: In Progress
# Agent: Competitive Intelligence Analyst ## Task: Explain how these changes may affect our bank's retail offerings or strategy. # Agent: Competitive Intelligence Analyst ## Final Answer: The recent changes in the retail banking landscape, particularly the Bank of Canada's interest rate adjustments and Scotiabank's product updates, present several strategic implications and opportunities for our bank. 1. **Interest Rate Strategy**: With the current interest rate at 2.75% and anticipated cuts in 2026, our bank should consider adjusting our interest rate offerings on loans and savings products. Lowering rates on loans could stimulate borrowing, particularly in a competitive market where other banks may also reduce rates. Conversely, we should evaluate our savings account rates to ensure they remain attractive to customers, balancing the need to maintain margins while being competitive. 2. **Capital Management**: Scotiabank's redemption of substantial capital notes indicates a strategic move to strengthen their balance sheet. Our bank should assess our capital structure and consider similar actions if it aligns with our financial strategy. This could enhance our creditworthiness and provide us with more flexibility to invest in growth opportunities or absorb potential losses during economic downturns. 3. **Economic Resilience**: The expectation for Canadian banks to build reserves in light of economic uncertainties suggests a cautious approach is warranted. Our bank should prioritize strengthening our reserve ratios to mitigate risks associated with potential economic downturns, such as trade wars. This could involve reassessing our lending criteria and focusing on high-quality borrowers to maintain asset quality. 4. **Wealth Management Focus**: The robust performance reported by major banks in wealth management indicates a growing opportunity in this sector. Our bank should consider enhancing our wealth management services, potentially through partnerships or technology investments, to capture a share of this lucrative market. This could involve developing tailored investment products or advisory services that cater to the evolving needs of our clients. 5. **Customer Engagement and Communication**: Given the cautious economic outlook, it is crucial for our bank to maintain open lines of communication with our customers. We should proactively engage with clients to provide insights on how interest rate changes may affect their financial decisions. This could enhance customer loyalty and position our bank as a trusted advisor during uncertain times. In summary, the current competitive landscape necessitates a strategic reassessment of our interest rate offerings, capital management, reserve building, wealth management focus, and customer engagement strategies. By proactively addressing these areas, our bank can capitalize on emerging opportunities while mitigating potential risks associated with the evolving economic environment.
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed └── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 Status: Executing Task... └── 🤖 Agent: Competitive Intelligence Analyst Status: ✅ Completed
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed └── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 Assigned to: Competitive Intelligence Analyst Status: ✅ Completed └── 🤖 Agent: Competitive Intelligence Analyst Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 │ │ Agent: Competitive Intelligence Analyst │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed ├── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 │ Assigned to: Competitive Intelligence Analyst │ Status: ✅ Completed │ └── 🤖 Agent: Competitive Intelligence Analyst │ Status: ✅ Completed └── 📋 Task: 49ff0cc0-e2ab-43b5-b8aa-abb8298d73ce Status: Executing Task...
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed ├── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 │ Assigned to: Competitive Intelligence Analyst │ Status: ✅ Completed │ └── 🤖 Agent: Competitive Intelligence Analyst │ Status: ✅ Completed └── 📋 Task: 49ff0cc0-e2ab-43b5-b8aa-abb8298d73ce Status: Executing Task... └── 🤖 Agent: Report Generator Status: In Progress
# Agent: Report Generator ## Task: Create a short internal report with search highlights, summaries, and analysis. # Agent: Report Generator ## Final Answer: # Internal Summary Report: Strategic Implications and Opportunities in Retail Banking ## Findings 1. **Interest Rate Strategy** - Current interest rate stands at 2.75%, with anticipated cuts projected for 2026. - Recommendation: Adjust interest rates on loans to stimulate borrowing while ensuring savings account rates remain competitive to attract customers. 2. **Capital Management** - Scotiabank's recent redemption of capital notes indicates a strategy to strengthen their balance sheet. - Recommendation: Assess our capital structure for potential similar actions to enhance creditworthiness and flexibility for growth investments. 3. **Economic Resilience** - Canadian banks are expected to build reserves due to economic uncertainties. - Recommendation: Strengthen reserve ratios and reassess lending criteria to focus on high-quality borrowers, mitigating risks from potential economic downturns. 4. **Wealth Management Focus** - Major banks report robust performance in wealth management, indicating growth opportunities. - Recommendation: Enhance wealth management services through partnerships or technology investments to capture market share, including tailored investment products and advisory services. 5. **Customer Engagement and Communication** - Maintaining open communication with customers is crucial in a cautious economic outlook. - Recommendation: Proactively engage clients with insights on interest rate changes to enhance loyalty and position our bank as a trusted advisor. ## Recommendations - **Interest Rate Adjustments**: Implement a strategic review of our interest rate offerings to remain competitive in the market, particularly focusing on loan rates to encourage borrowing while ensuring savings rates attract deposits. - **Capital Structure Review**: Conduct a thorough assessment of our capital management strategy, considering the redemption of capital notes to strengthen our balance sheet and improve our financial flexibility. - **Reserve Building Strategy**: Prioritize the enhancement of our reserve ratios to prepare for economic uncertainties, focusing on high-quality lending to maintain asset quality. - **Investment in Wealth Management**: Explore opportunities to expand our wealth management services, potentially through strategic partnerships or technology enhancements, to meet the growing demand in this sector. - **Enhanced Customer Communication**: Develop a proactive communication strategy to keep customers informed about interest rate changes and their implications, fostering trust and loyalty. In conclusion, the evolving retail banking landscape presents both challenges and opportunities. By strategically addressing interest rates, capital management, reserves, wealth management, and customer engagement, our bank can position itself for success in a competitive environment while mitigating risks associated with economic fluctuations.
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed ├── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 │ Assigned to: Competitive Intelligence Analyst │ Status: ✅ Completed │ └── 🤖 Agent: Competitive Intelligence Analyst │ Status: ✅ Completed └── 📋 Task: 49ff0cc0-e2ab-43b5-b8aa-abb8298d73ce Status: Executing Task... └── 🤖 Agent: Report Generator Status: ✅ Completed
🚀 Crew: crew ├── 📋 Task: f7fba45b-2450-4125-abb0-f6979e46e344 │ Assigned to: Search Agent │ Status: ✅ Completed │ └── 🤖 Agent: Search Agent │ Status: ✅ Completed ├── 📋 Task: 985e1e8e-615e-4766-9195-3da0c4d38c82 │ Assigned to: Summarizer Agent │ Status: ✅ Completed │ └── 🤖 Agent: Summarizer Agent │ Status: ✅ Completed ├── 📋 Task: 22a95991-621c-4f7d-b4fe-ef6a4b2b6675 │ Assigned to: Competitive Intelligence Analyst │ Status: ✅ Completed │ └── 🤖 Agent: Competitive Intelligence Analyst │ Status: ✅ Completed └── 📋 Task: 49ff0cc0-e2ab-43b5-b8aa-abb8298d73ce Assigned to: Report Generator Status: ✅ Completed └── 🤖 Agent: Report Generator Status: ✅ Completed
╭──────────────────────────────────────────────── Task Completion ────────────────────────────────────────────────╮ │ │ │ Task Completed │ │ Name: 49ff0cc0-e2ab-43b5-b8aa-abb8298d73ce │ │ Agent: Report Generator │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭──────────────────────────────────────────────── Crew Completion ────────────────────────────────────────────────╮ │ │ │ Crew Execution Completed │ │ Name: crew │ │ ID: 30dd2315-3249-4efe-bc7c-665276b3c1e5 │ │ │ │ │ ╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
In [56]:
print(result)
# Internal Summary Report: Strategic Implications of Recent Banking Changes ## Findings 1. **Interest Rate Environment** The Bank of Canada has reduced interest rates to 2.75%, with expectations of further cuts. This shift necessitates a reassessment of our pricing strategy for loans and deposits. Lower interest rates typically compress margins on lending products, creating an opportunity for our bank to enhance its competitive positioning by offering attractive loan products. This could allow us to capture market share from competitors, particularly Scotiabank, as they adjust their offerings in response to these rate changes. 2. **Product Innovation and Differentiation** Scotiabank's recent redemption of significant capital notes indicates a strategic focus on strengthening its balance sheet, which may lead to new product offerings or enhancements. Our bank should consider innovating our product suite to stand out in the market. Potential initiatives could include introducing flexible mortgage options or competitive savings accounts that cater to consumers seeking stability amid fluctuating interest rates. 3. **Wealth Management Focus** The strong performance of major banks in the wealth management sector highlights a growing opportunity for our bank. We should evaluate our current wealth management services and consider expanding or enhancing our offerings to attract high-net-worth clients. This could involve forming strategic partnerships or investing in technology to provide superior financial planning tools and investment options. 4. **Risk Management and Reserves** In light of tariff uncertainties, Canadian banks are expected to bolster their reserves. Our bank must also review its risk management strategies, which may include increasing capital reserves to mitigate potential impacts from economic fluctuations. Additionally, conducting stress tests on our portfolio against various economic scenarios will ensure we are prepared for potential downturns. 5. **Market Positioning and Communication** With ongoing discussions about trade wars affecting interest rates, it is crucial for our bank to proactively communicate our strategies and positioning to customers. Transparency regarding our risk management and adaptation to market conditions can enhance customer trust and loyalty, especially during uncertain times. ## Recommendations - **Adjust Pricing Strategy**: Reassess our loan and deposit pricing to remain competitive in the lower interest rate environment, potentially offering promotional rates to attract new customers. - **Innovate Product Offerings**: Develop new and differentiated products, such as flexible mortgage options and competitive savings accounts, to meet the evolving needs of consumers. - **Expand Wealth Management Services**: Invest in enhancing our wealth management offerings and consider partnerships that can provide advanced financial planning tools to attract high-net-worth clients. - **Strengthen Risk Management**: Increase capital reserves and conduct regular stress testing of our portfolio to ensure resilience against economic fluctuations. - **Enhance Communication Strategy**: Develop a proactive communication plan to inform customers about our strategies and how we are managing risks, thereby fostering trust and loyalty. ## Conclusion The recent changes in the banking landscape present both risks and opportunities for our bank. By strategically adjusting our offerings, focusing on innovation, enhancing our wealth management services, and strengthening our risk management practices, we can position ourselves favorably in a competitive landscape while addressing the evolving needs of our customers.
Send Email
Below is a function that sends the result of CrewAI agent chain via email:
In [57]:
import smtplib
from email.mime.text import MIMEText
from email.mime.multipart import MIMEMultipart
def send_email_report(subject, report_text, to_emails):
msg = MIMEMultipart()
msg["From"] = EMAIL_ADDRESS
msg["To"] = ", ".join(to_emails)
msg["Subject"] = subject
# Attach the text version of the report
msg.attach(MIMEText(report_text, "plain"))
# Send via SMTP
with smtplib.SMTP_SSL("smtp.gmail.com", 465) as server:
server.login(EMAIL_ADDRESS, EMAIL_PASSWORD)
server.send_message(msg)
print(f"✅ Email sent to: {', '.join(to_emails)}")
In [ ]:
In [73]:
# Email config
EMAIL_ADDRESS = "mrezvandehy@gmail.com"
EMAIL_PASSWORD = ".... .... .... ...."
recipients = ["mehdi_rezvandehy@ucalgary.ca", "m_rezvandehy@yahoo.com", "mrezvandehy@gmail.com"]
# Send the final report
send_email_report(
subject="📊 Weekly Competitive Intelligence Report",
report_text=str(result),
to_emails=recipients
)
✅ Email sent to: mehdi_rezvandehy@ucalgary.ca, m_rezvandehy@yahoo.com, mrezvandehy@gmail.com
In [ ]: