This open-source course teaches how to build an AI-powered game simulation engine that impersonates historical philosophers. It targets ML/AI, Data, and Software Engineers looking to create production-ready agentic applications beyond notebook tutorials, offering a hands-on approach to combining game development with LLM and RAG technologies.

How It Works

The course focuses on building an agentic RAG system using LangGraph and LangChain to create AI agents that embody philosophers like Plato and Aristotle. It emphasizes production-ready architecture, including RESTful API deployment with FastAPI and WebSockets, and integrates with tools like Groq for high-speed inference and MongoDB for agent memory. The approach prioritizes practical application of LLMOps and software engineering best practices.

Quick Start & Requirements

• Install/Run: Clone the repository and follow instructions in INSTALL_AND_USAGE.md.
• Prerequisites: Python (beginner), ML/LLM/RAG (beginner). Requires a modern laptop/PC. Uses Groq API (free tier available) and optionally OpenAI API (~$1 cost).
• Resources: Course materials include videos and articles. The project structure includes philoagents-api (Python backend) and philoagents-ui (Node frontend), with the course focusing on the API.
• Links: code, videos, articles

Highlighted Details

• Build AI agents that authentically impersonate historical philosophers.
• Architect and implement production-ready RAG, LLM, and LLMOps systems from scratch.
• Expose agents as RESTful APIs with real-time communication via WebSockets.
• Integrate with industry tools: Groq, MongoDB, Opik, LangGraph, LangChain, FastAPI, Docker.

Maintenance & Community

• Collaborations with MongoDB, Opik, and Groq.
• Core Contributors: Paul Iusztin, Miguel Otero Pedrido.
• Stay updated via newsletters from The Neural Maze and Decoding ML.

Licensing & Compatibility

• MIT License. Permissive for commercial use and closed-source linking.

Limitations & Caveats

Modules 5 and 6 (LLMOps and multi-agent simulation) are marked as Work In Progress (WIP). The course focuses on the backend API; the UI is provided but not the primary focus of instruction.