// series · 7 parts
Agentic AI in Python: Zero to Production
Learn how to build a working agentic AI app in Python with LangGraph. Part 1 covers tools, StateGraph, and memory — step by step from scratch.
SK
Sukhveer Kaur
Updated Jun 20, 2026
~53 min
Total read
Intermediate
Level
7 parts published
// the path
01
Build an Agentic AI App in Python: Zero to Production (Part 1)
Learn how to build a working agentic AI app in Python with LangGraph. Part 1 covers tools, StateGraph, and memory — step by step from scratch.
9 min
✓ Published02
Build an Agentic AI App in Python: FastAPI, Docker & Deploy to Production (Part 2)
Wrap your LangGraph agent in FastAPI, Dockerize it, and deploy to a cloud VPS. Part 2 of the Agentic AI Python series — zero to production.
9 min
✓ Published03
Build an Agentic AI App in Python: Multi-Agent Systems (Part 3)
Build multi-agent systems in Python with LangGraph: a supervisor routing search, summarise & fact-check workers, plus loop guards that save real money.
8 min
✓ Published04
Build an Agentic AI App in Python: AI Agent Memory (Part 4)
Add real AI agent memory in Python — a LangGraph Store that recalls users across threads, semantic search with embeddings, and Redis for production.
7 min
✓ Published05
Build an Agentic AI App in Python: MCP Client (Part 5)
Build an MCP client in Python so your agent calls real tools — connect a LangGraph agent to an authenticated FastMCP server with langchain-mcp-adapters.
7 min
✓ Published06
AI Agent Observability & Evals in Python (Part 6)
Add AI agent observability and evals in Python — trace every agent run with Pydantic Logfire and score it against a test set using pydantic-evals.
7 min
✓ Published07
AI Agent Evals in CI: Block Bad PRs with GitHub ActionsLATEST
Run AI agent evals in CI: turn your pydantic-evals suite into a test, gate it with GitHub Actions, and block any pull request that drops the agent's score.
6 min
✓ Published