Building Agentic Systems

A hands-on Python course that builds a working, provider-agnostic agent framework from scratch — the agent loop, tools, memory, planning, multi-agent orchestration, guardrails, and evaluation. Every lesson runs for real against a deterministic mock model (zero cost, no API key), and the capstone shows how to plug in a real LLM provider. The codebase grows lesson by lesson in one public repo.

1 of 9 published