Learning CrewAI - Multi-Agent AI System Projects

<div align="center"> <img src="https://raw.githubusercontent.com/Tarikul-Islam-Anik/Animated-Fluent-Emojis/master/Emojis/People/People%20Hugging.png" alt="Multi-Agent Collaboration" width="200"/>

</div>

---

<div align="center">

Intelligent Multi-Agent Collaboration | Production-Ready Systems | Real-World Applications

</div>

---

A comprehensive collection of enterprise-grade, multi-agent AI systems built with CrewAI. Each project demonstrates collaborative AI agents solving complex real-world problems through intelligent task orchestration and domain-specific expertise.

---

Overview

<div align="center">

graph LR
    A[Stock Researcher] --> E[Multi-Agent System]
    B[Cricket Analytics] --> E
    C[Budget Knowledge] --> E
    D[Research Writer] --> E
    E --> F[Intelligent Solutions]

</div>

This workspace contains multiple CrewAI projects that showcase advanced multi-agent AI capabilities:

• Stock Researcher: Financial analysis with S&P 500 benchmarking and investment recommendations
• Cricket Performance Analysis: Sports analytics with player performance metrics and visualizations
• Union Budget Knowledge: Government budget analysis and policy query system
• Research Writer Crew: Automated research and article generation

Quick Start

1. Environment Setup

# Navigate to workspace root
cd /Users/shinde/Desktop/Talbot/learning_crewai

# Activate shared virtual environment
source venv/bin/activate

# Create .env file with API keys
cp .env.example .env
# Edit .env and add your API keys:
# OPENAI_API_KEY=sk-...
# SERPER_API_KEY=...

2. Run a Project

# Stock Researcher
cd stock_researcher/src
python main.py "AAPL"

# Cricket Performance Analysis
cd cricketer_performance_analysis/src
python main.py "Virat Kohli" "2023"

# Union Budget Knowledge
cd union_budget_knowledge
PYTHONPATH=src python -c 'from union_budget_knowledge.main import kickoff; kickoff()'

# Research Writer Crew
cd research_writer_crew/src
python main.py

---

Projects

1. Stock Researcher

Advanced financial analysis system with S&P 500 benchmarking

<div align="center">

</div>

• Agents: 6 specialized financial analysts
• Capabilities:
  • Real-time stock data retrieval
  • Comparative S&P 500 analysis
  • Investment recommendations (BUY/HOLD/SELL)
  • Professional visualization and reporting

→ Full Documentation

Quick Run:

cd stock_researcher/src
python main.py "AAPL"  # Analyze Apple stock

---

2. Cricket Performance Analysis

Sports analytics and player performance evaluation system

<div align="center">

</div>

• Agents: 3 specialized sports analysts
• Capabilities:
  • Cricket player statistics aggregation
  • Performance metric calculations
  • Format-wise analysis (Test/ODI/T20)
  • Professional visualization and reporting

→ Full Documentation

Quick Run:

cd cricketer_performance_analysis/src
python main.py "Virat Kohli" "2023"

---

3. Union Budget Knowledge

Government budget analysis and policy query system

<div align="center">

</div>

• Agents: Multi-agent budget analysis system
• Capabilities:
  • PDF document processing and indexing
  • Natural language budget queries
  • Policy analysis and insights
  • Interactive Q&A interface

→ Full Documentation

Quick Run:

cd union_budget_knowledge
PYTHONPATH=src python -c 'from union_budget_knowledge.main import kickoff; kickoff()'

---

4. Research Writer Crew

Automated research and article generation system

<div align="center">

</div>

• Agents: Research and writing specialists
• Capabilities:
  • Topic research and data gathering
  • Article outline generation
  • Content writing and editing
  • SEO optimization

→ Full Documentation

Quick Run:

cd research_writer_crew/src
python main.py

---

Shared Resources

Documentation

Location: docs/

Comprehensive learning resources and guides:

• Complete A-Z Guide to CrewAI - The ultimate 6,000+ line reference covering:
  • All CrewAI concepts from Agents to Zero-to-Production
  • Architecture patterns and best practices
  • Prompt engineering techniques
  • Production deployment strategies
  • Real-world code examples
  • Debugging and monitoring

Virtual Environment

Location: venv/

All projects use a shared Python virtual environment for efficient dependency management.

Setup:

# One-time setup (already done)
python3 -m venv venv
source venv/bin/activate
pip install crewai[tools] python-dotenv pypdf2 litellm

# Subsequent uses
source venv/bin/activate

Environment Configuration

Location: .env

Centralized API key management for all projects:

# Required for all projects
OPENAI_API_KEY=sk-your_openai_key
SERPER_API_KEY=your_serper_key

# Optional project-specific settings
LOG_LEVEL=INFO
DEBUG_MODE=false

Project Structure

learning_crewai/
├── .env                              # Shared environment variables
├── .gitignore                        # Git ignore rules
├── venv/                             # Shared virtual environment
│   ├── bin/
│   ├── lib/
│   └── ...
│
├── docs/                             # Documentation
│   └── crew_ai_context.md           # Complete A-Z Guide to CrewAI
│
├── stock_researcher/                 # Financial analysis project
│   ├── src/
│   │   ├── main.py
│   │   ├── crew.py
│   │   └── config/
│   │       ├── agents.yaml
│   │       └── tasks.yaml
│   └── README.md
│
├── cricketer_performance_analysis/   # Sports analytics project
│   ├── src/
│   │   ├── main.py
│   │   ├── crew_setup.py
│   │   └── config/
│   │       ├── agents.yaml
│   │       └── tasks.yaml
│   └── README.md
│
├── union_budget_knowledge/           # Budget analysis project
│   ├── src/
│   │   ├── main.py
│   │   ├── crews/
│   │   │   └── budget_response_crew/
│   │   │       ├── budget_crew.py
│   │   │       └── config/
│   │   │           ├── agents.yaml
│   │   │           └── tasks.yaml
│   │   └── tools/
│   ├── Union_Budget_Analysis-2023-24.pdf
│   └── README.md
│
├── research_writer_crew/             # Research & writing project
│   ├── src/
│   │   ├── main.py
│   │   ├── crew_setup.py
│   │   └── config/
│   │       ├── agents.yaml
│   │       └── tasks.yaml
│   └── README.md
│
└── README.md                         # This file

Prerequisites

System Requirements

• OS: macOS, Linux, or Windows
• Python: 3.10, 3.11, 3.12, or 3.13
• Memory: Minimum 4GB RAM (8GB recommended)
• Disk Space: 5GB for dependencies

API Keys Required

1. OpenAI API Key

   • Get at: https://platform.openai.com/api-keys
   • Cost: Pay-as-you-go (free tier available)
   • Models: GPT-4, GPT-3.5-turbo

2. Serper API Key (for web search projects)

   • Get at: https://serper.dev
   • Cost: Free tier (100 searches/month), paid plans from $5/month
   • Used by: Stock Researcher, Cricket Analysis

Setup Instructions

1. Initial Setup (One-time)

# Clone or navigate to the workspace
cd /Users/shinde/Desktop/Talbot/learning_crewai

# Create virtual environment (if not exists)
python3 -m venv venv

# Activate virtual environment
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt
# Or install manually:
pip install crewai[tools]>=0.86.0 python-dotenv pypdf2 litellm

# Create .env file
cp .env.example .env
# Edit .env with your API keys
nano .env

2. For Each Session

# Activate virtual environment
cd /Users/shinde/Desktop/Talbot/learning_crewai
source venv/bin/activate

# Navigate to desired project
cd <project_name>

# Run the project
python main.py [arguments]

3. Configuration

Update the .env file in the workspace root:

# File: /Users/shinde/Desktop/Talbot/learning_crewai/.env

# OpenAI Configuration (Required)
OPENAI_API_KEY=sk-your_key_here
OPENAI_MODEL_NAME=gpt-4  # or gpt-3.5-turbo

# Serper Configuration (Required for Stock & Cricket projects)
SERPER_API_KEY=your_serper_key_here

# Logging Configuration (Optional)
LOG_LEVEL=INFO  # DEBUG, INFO, WARNING, ERROR
DEBUG_MODE=false

# Project Settings (Optional)
STOCK_SYMBOL=AAPL
CRICKETER_NAME=Virat Kohli
YEAR=2023

Architecture Overview

<div align="center">

graph TB
    subgraph Workspace["Learning CrewAI Workspace"]
        subgraph Resources["Shared Resources"]
            ENV[.env Configuration]
            VENV[Python Virtual Environment]
            DOCS[Documentation Hub]
        end
        
        subgraph Projects["Multi-Agent Projects"]
            STOCK[Stock Researcher<br/>Financial Analysis]
            CRICKET[Cricket Analytics<br/>Sports Intelligence]
            BUDGET[Budget Knowledge<br/>Policy Analysis]
            WRITER[Research Writer<br/>Content Creation]
        end
        
        subgraph Framework["CrewAI Framework"]
            AGENTS[Specialized Agents]
            TASKS[Orchestrated Tasks]
            TOOLS[Integrated Tools]
        end
        
        subgraph External["External Services"]
            OPENAI[OpenAI GPT-4]
            SERPER[Serper Search API]
            DATA[Data Sources]
        end
    end
    
    ENV --> Projects
    VENV --> Projects
    DOCS --> Projects
    
    STOCK --> Framework
    CRICKET --> Framework
    BUDGET --> Framework
    WRITER --> Framework
    
    Framework --> External
    
    AGENTS -.-> TASKS
    TASKS -.-> TOOLS
    
    style Workspace fill:#f9f9f9,stroke:#333,stroke-width:2px
    style Resources fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
    style Projects fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    style Framework fill:#fff3e0,stroke:#f57c00,stroke-width:2px
    style External fill:#e8f5e9,stroke:#388e3c,stroke-width:2px

</div>

System Flow

<div align="center">

sequenceDiagram
    participant User
    participant Project
    participant CrewAI
    participant Agents
    participant LLM
    participant Tools
    
    User->>Project: Execute Command
    Project->>CrewAI: Initialize Crew
    CrewAI->>Agents: Assign Tasks
    
    loop Task Execution
        Agents->>LLM: Process Information
        LLM-->>Agents: Generate Response
        Agents->>Tools: Use External Tools
        Tools-->>Agents: Return Data
    end
    
    Agents->>CrewAI: Complete Tasks
    CrewAI->>Project: Aggregate Results
    Project->>User: Deliver Output

</div>

---

Features Across All Projects

<div align="center">

| Architecture | Configuration | Integration | Output | |:---:|:---:|:---:|:---:| | Multi-agent system | YAML-based | Real-time APIs | Professional reports | | Specialized roles | Environment vars | OpenAI & Serper | Structured data | | Task orchestration | Modular design | External tools | Rich formatting |

</div>

Common Capabilities

• Multi-agent architecture with specialized roles
• YAML-based configuration for agents and tasks
• Real-time API integration (OpenAI, Serper)
• Comprehensive logging and error handling
• Professional output formatting
• Modular and extensible design
• Environment-based configuration

Project-Specific Features

| Project | Key Features | |---------|-------------| | Stock Researcher | Real-time stock data, benchmark analysis, investment ratings | | Cricket Analysis | Player statistics, format analysis, performance visualization | | Budget Knowledge | PDF processing, policy analysis, interactive Q&A | | Research Writer | Topic research, outline generation, content writing |

Usage Examples

Stock Researcher

cd stock_researcher/src
python main.py "AAPL"      # Apple
python main.py "GOOGL"     # Google
python main.py "MSFT"      # Microsoft

Output: Investment recommendations, S&P 500 comparison, analysis files

Cricket Performance Analysis

cd cricketer_performance_analysis/src
python main.py "Virat Kohli" "2023"
python main.py "Babar Azam" "2022"
python main.py "Steve Smith" "2021"

Output: Performance statistics, visualizations, analysis reports

Union Budget Knowledge

cd union_budget_knowledge
PYTHONPATH=src python -c 'from union_budget_knowledge.main import kickoff; kickoff()'

Interactive Mode: Ask budget-related questions, get instant analysis

Research Writer Crew

cd research_writer_crew/src
python main.py

Interactive Mode: Enter topic, get comprehensive research and article

Customization Guide

Adding a New Project

1. Create project structure:

mkdir my_new_project
cd my_new_project
mkdir -p src/my_new_project/config

1. Create configuration files:

# agents.yaml
touch src/my_new_project/config/agents.yaml

# tasks.yaml
touch src/my_new_project/config/tasks.yaml

1. Create crew file:

touch src/my_new_project/crew_setup.py

1. Create main entry point:

touch src/my_new_project/main.py

1. Update root README to list the new project

Extending Existing Projects

Each project can be extended by:

• Adding new agents in config/agents.yaml
• Adding new tasks in config/tasks.yaml
• Creating custom tools in the tools/ directory
• Modifying crew orchestration in crew.py or crew_setup.py

Troubleshooting

Common Issues

"Virtual environment not activated"

source /Users/shinde/Desktop/Talbot/learning_crewai/venv/bin/activate

"ModuleNotFoundError"

# Reinstall dependencies
pip install crewai[tools] python-dotenv

"API Key not found"

# Verify .env file exists and contains keys
cat /Users/shinde/Desktop/Talbot/learning_crewai/.env
export OPENAI_API_KEY=your_key

"Disk space full"

# Free up space
rm -rf ~/.cache/pip
pip cache purge

Debug Mode

Enable detailed logging:

export LOG_LEVEL=DEBUG
export DEBUG_MODE=true
python main.py

Performance & Optimization

API Rate Limiting

• OpenAI: Monitor at https://platform.openai.com/account/usage
• Serper: Check quota at https://serper.dev/dashboard

Cost Optimization

• Use GPT-3.5-turbo for cost savings
• Cache agent responses when possible
• Batch process queries

Execution Times

• Simple queries: 2-5 seconds
• Complex analysis: 10-30 seconds
• Large document processing: 1-2 minutes

Security Best Practices

1. Never commit .env file
2. Rotate API keys regularly
3. Monitor API usage for unauthorized access
4. Use environment variables only, never hardcode keys
5. Keep dependencies updated

Contributing

Development Workflow

# Create feature branch
git checkout -b feature/new-feature

# Make changes
git add .

# Commit with descriptive message
git commit -m "feat: description of changes"

# Push to repository
git push origin feature/new-feature

Code Standards

• Follow PEP 8 style guide
• Add type hints to functions
• Write comprehensive docstrings
• Add error handling
• Update README for user-facing changes

Monitoring & Metrics

What to Track

• API usage and costs
• Agent performance (response time, accuracy)
• Error rates and failures
• User query patterns

View Logs

# Recent logs
tail -f <project>/run_output.txt

# Search for errors
grep -i error <project>/run_output.txt

Support & Resources

• CrewAI Docs: https://docs.crewai.com/
• OpenAI API: https://platform.openai.com/docs/
• Serper API: https://serper.dev/docs
• Python Docs: https://docs.python.org/3/

Roadmap

Future Enhancements

• [ ] Database integration for persistent storage
• [ ] Web UI dashboard
• [ ] Advanced caching layer
• [ ] Multi-language support
• [ ] Real-time streaming responses
• [ ] Analytics dashboard

License

This project collection is provided as-is for educational and commercial use.

Version History

| Version | Date | Status | |---------|------|--------| | 1.0.0 | Dec 6, 2025 | Production Ready | | 0.9.0 | Dec 5, 2025 | Beta | | 0.1.0 | Dec 4, 2025 | Alpha |

Team & Contact

Maintained By: Shinde Vinayak Rao Patil Last Updated: December 6, 2025

---

Note: For project-specific documentation, refer to individual project READMEs linked above.