Algol Lab's Practical AI 🧪

A comprehensive, hands-on workshop designed to teach mid-to-senior engineers the fundamentals of building AI applications.

Theme: "Practical AI Corp" — You will build internal tools for HR policies (RAG) and performance evaluations (Agents).

📚 Workshop Content

Lab 1: Company Policy Assistant (RAG)

Build a Retrieval-Augmented Generation system to query internal PDF policies.

• Key Concepts: Data Ingestion, Vector Stores (Qdrant), Chunking strategies, Citation handling.
• Stack: LlamaIndex, Qdrant, Gradio.
• Start Lab 1

Lab 2: Performance Review Assistant (ReAct Agents)

Build an Agent that can read files, reason about goals, and act on them.

• Key Concepts: The ReAct Loop (Reason -> Act -> Observe), Tool use, Function Calling.
• Stack: LlamaIndex (Agent), Rich (CLI).
• Start Lab 2

📖 Documentation

We believe in understanding why, not just how.

• Core Theory: Deep dive into LLMs, Embeddings, RAG, and Agents.
• Setup Guide: Detailed installation instructions (Python, Docker, uv).
• Lab 1 Guide: RAG implementation details.
• Lab 2 Guide: Agent implementation details.

⚡ Quick Start

For detailed prerequisites and troubleshooting, see the Setup Guide.

1. Clone & Install

   git clone https://github.com/algollabs/practical-ai-lab.git
   cd practical-ai-lab
   uv sync

2. Configure

   cp .env.example .env
   # Add your OpenAI API Key in .env

3. Run Infrastructure

   docker-compose up -d

4. Run Labs

   # Lab 1 (RAG UI)
   uv run python -m lab1_rag.app
   
   # Lab 2 (Agent CLI)
   uv run python -m lab2_agents.cli

---

Built with ❤️ by Zaid Amireh for the tech community.