Learn to Build AI Agents

Free, research-backed tutorials. Every example runs. No accounts required for the local-first path.

The agentic-AI space is 90% hype. This hub is the other 10%: a curriculum that teaches you to build agents, grounded in what the ecosystem actually uses — ranked by GitHub stars, not Twitter trends.

Full codebase + all tutorials: github.com/CodeGero/ai-engineering-hub (MIT licensed).

Run AI Locally with Ollama

Free, private inference on your machine. No API key, no cloud.

Ollama · 176k★

Build Your First Agent (From Scratch)

An ~80-line ReAct loop in plain Python. Understand the mechanism, no frameworks.

Foundations

Production Agents with LangGraph

Stateful graphs, memory, human-in-the-loop.

LangChain 141k★

Multi-Agent Systems with AutoGen

Agent teams that generate, then critique each other.

AutoGen 60k★

MCP: Build a Tool Server

The new standard for giving agents real tools. Write once, use everywhere.

MCP 88k★

RAG with LlamaIndex

Ground agents in your own documents. Stop the hallucination.

LlamaIndex 51k★

Prompt Engineering That Works

The skill under every agent. Specificity, few-shot, negative constraints.

Foundations

Ship Agents to Production

Eval, guardrails, cost control, observability.

Ops

Agent Skills

Package capabilities for reuse. The 160k★ pattern.

Skills 160k★

Want the tools, not the tutorials? Browse the BTC store →