I’ve been building AI systems for over twenty years. From natural language understanding at Evi Technologies, to helping launch Alexa at Amazon, to running my own AI company. I’ve seen a lot of hype cycles come and go.
This one is different.
AI agents are genuinely changing how engineering teams work. Not in the “we added a chatbot to our app” sense — in the “a team of five can now do what used to take fifty” sense. But most of the conversation around this is either breathless hype or dismissive scepticism. Neither is useful.
Why write about it?
Two reasons.
First, I think clearly by writing clearly. The act of explaining something well forces you to understand it properly. When I’m helping engineering teams adopt AI agents, the hard part isn’t the technology — it’s knowing which patterns actually work in production and which only work in demos.
Second, the teams I work with keep asking the same questions. How do we start? What tools should we use? How do we avoid the pitfalls? Writing those answers down means more people benefit.
What to expect
I’ll write about AI agents, agentic workflows, and the practical reality of adopting AI in engineering teams. Some posts will be technical. Some will be about the human side — how teams change, what leaders need to understand, where things go wrong.
All of it will be honest. If something doesn’t work, I’ll say so. If I don’t know, I’ll say that too.
No publishing schedule. Just writing when I have something worth saying.