Rudnik-Ilia/Agents-Sandbox

A comprehensive educational lab for AI agent architectures built on LangChain & Ollama. Compare diverse patterns: simple chat memory, advanced tool calling (ReAct), and multiple RAG backends (Chroma, LlamaIndex, NumPy, Rerank). Study intelligent routing (Adapt/Semant) and Agentic workflows via a unified CLI. See how LLMs actually decide what to do!

1 best points · 0 comments · 1 Hacker News post · 2 GitHub stars · View on GitHub

Hacker News discussions

• An educational lab of AI agent architectures (1 points, 0 comments, by ilia53, 2026-07-11)