Welcome to the Agentic AI Workshop! This repository contains hands-on Jupyter notebooks demonstrating how to build intelligent AI agents using LangChain, LangGraph, and modern LLMs.
This workshop is designed to teach you how to build agentic AI systems through two hands-on Jupyter Notebooks:
Learn how to build an agentic workflow that generates realistic synthetic HR databases based on natural language company descriptions.
Key Features:
- Convert user descriptions into structured company specifications using LLMs
- Generate demographic ratios based on company characteristics
- Create complete HR database schema with business units, departments, jobs, and employees
- Parallel processing for efficient data generation
- Automatic compensation calculation and employee education assignment
Build an intelligent agent that can query and analyze HR databases using natural language.
Key Features:
- Natural language to SQL query translation
- Automatic SQL execution and data retrieval
- Statistical analysis and calculations
- Data visualization with Plotly
- Structured responses with summaries, datasets, and charts
Click on the badges below to open the notebooks directly in Google Colab:
-
HR Synthetic Database Generator:
-
People Analytics Agent:
All dependencies will be installed automatically when you run the first cell of each notebook.
- Python 3.12 or higher
- uv package manager (recommended) or pip
-
Clone the repository:
git clone https://github.com/dougtrajano/agentic-ai-workshop.git cd agentic-ai-workshop -
Install dependencies using uv (recommended):
# Install with development dependencies uv sync --group devOr using pip:
# Create virtual environment python -m venv .venv source .venv/bin/activate # On Windows: .venv\Scripts\activate # Install dependencies pip install -e ".[dev]"
-
Start JupyterLab:
# With uv uv run jupyter lab # Or with activated virtual environment jupyter lab
-
Set up your API keys:
Both notebooks require API keys for LLM access (e.g., Google Gemini, OpenAI). You'll be prompted to enter these when running the notebooks, or you can set them as environment variables.
- Agentic Workflows: Build multi-step AI workflows with LangGraph
- Tool Use: Create and integrate custom tools for AI agents
- LLM Integration: Work with modern LLMs like Google Gemini
- Database Operations: Generate and query SQL databases with AI
- Data Visualization: Create interactive charts and dashboards
- Production Patterns: Implement error handling, streaming, and structured outputs
├── hr_synthetic_database.ipynb # Database generation agent notebook
├── people_analytics_agent.ipynb # People analytics query agent notebook
├── pyproject.toml # Project dependencies
├── LICENSE # Apache 2.0 license
└── README.md # This file
This workshop uses several key libraries for building agentic AI systems. For the complete list of dependencies and their versions, please refer to the pyproject.toml file.
Main libraries include:
- LangChain, LangGraph, LangChain Community: Agent orchestration and workflow management
- Google Generative AI & OpenAI: LLM integration (Gemini and GPT models)
- Datasets & Hugging Face Hub: Data handling and storage
- Pandas, NumPy, Plotly: Data analysis and visualization
- SQLAlchemy, DuckDB: Database operations and querying
- Pydantic: Data validation and settings management
- Faker: Synthetic data generation
- Chainlit: Interactive chat interfaces
- pyngrok: Tunneling for public access
All dependencies are automatically installed when running the notebooks in Google Colab or when using the local installation methods above.
Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
If you have questions or run into issues:
- Check the notebook documentation and comments
- Review the Issues page
- Open a new issue with detailed information about your problem
Happy Learning! 🚀
Built with ❤️ by ADP Brazil Labs