Agentic AI
Most agentic AI demos are impressive. Most production deployments are not. This is where I write about closing that gap — evaluating agents, controlling cost, designing human-in-the-loop operations, and building the data foundations agents actually need. Everything here comes from a decade of shipping production AI systems at a global cloud provider.
Start here
The four pieces that define my production-first view of agentic AI:
- The Architect’s Dilemma: A Defensible Framework for Agentic ROI
- When Do Multi-Agent AI Systems Actually Scale?
- AFK AI Coding with “Ralph”: Let Your AI Code While You’re Away
- RAG in Production: What the Demos Don’t Tell You
Foundations
The building blocks — models, embeddings, vector stores, and orchestration frameworks:
- Agentic AI Boot Camp: A Hands-On Journey
- LangChain and LangGraph
- Understanding the Brains Behind Generative AI: LLM
- Understanding Embeddings
- Vector Databases
- Beyond the AI Hype: The 3 Breakthroughs Needed to Reach Next-Gen Intelligence
- Machine Learning Basics and Foundations
From the field
How these ideas survive contact with production — real systems, real constraints:
The Agentic Enterprise (newsletter)
A monthly note on production AI, agentic systems, and the data architecture behind them — for technology leaders shipping AI that has to work. No fluff, no hype, unsubscribe anytime.
