Announcing "Agentic Engineering: Building Autonomous AI Systems with Python"
I’m excited to announce that my new book, Agentic Engineering: Building Autonomous AI Systems with Python, is now available on Amazon.
Over the past year, I’ve been writing a comprehensive technical guide to building production-grade autonomous AI systems. The book covers everything from the foundations of agentic reasoning to multi-agent orchestration, memory systems, observability, and production architecture patterns.
What’s in the book
The book is structured around 16 chapters covering the full lifecycle of building and running agentic systems:
- Reasoning frameworks — ReAct, extended thinking, hierarchical planning, and self-reflection
- Memory systems — short-term, long-term, episodic, and semantic memory patterns
- Multi-agent orchestration — hierarchical, peer-to-peer, and marketplace patterns
- Agentic RAG — active retrieval, query decomposition, and iterative refinement
- Observability — tracing agent thought processes with LangSmith and OpenTelemetry
- Guardrails and ethics — responsible AI patterns for production systems
- FinOps — token economics, prompt caching, and cost optimization at scale
- Framework deep dives — LangGraph, CrewAI, and Microsoft Agent Framework
- Production case studies — real deployments with quantified results
The book is aimed at software engineers and architects who are building — or planning to build — autonomous AI systems in production.
Why I wrote it
The field of agentic AI is moving fast. While there is no shortage of blog posts and tutorials, I found it hard to locate a single resource that covered the full picture: from the theoretical underpinnings of reasoning frameworks to the practical details of deploying multi-agent systems at scale, with real production metrics and verified code. This book is my attempt to fill that gap.
Get it
If you read it, I’d love to hear your feedback — and an Amazon review would be greatly appreciated.