AI Agents

Spec-driven development with AI works surprisingly well in isolation. You write a concise spec, let an agent generate the implementation, and get clean results on the first run. No endless prompt tweaking required. The illusion breaks the moment you bring this into a team. The question shifts immediately from “does this work?” to: “How does this hold up with five engineers on a shared repository over the next six months?” ...

Over the past months, a question has been making the rounds in engineering circles: Is anyone doing “triplet programming” — two humans and an AI agent? We are. Here is why. The Setup When we introduced GitHub Copilot as an agentic coding assistant, the default assumption was that each developer would work one-on-one with the agent. That is not what we ended up doing. Instead, we run sessions with two developers and one agent. ...

Over the past months, I have experimented extensively with using AI in software and system architecture. One pattern became increasingly clear: We try to use AI like a compiler. We assume that if we specify enough, the AI should be able to generate an entire system. But this approach fails consistently. Not because AI is “too dumb” — but because we are using the wrong model. In this article, I explain why full specification does not work with AI and which architectural model does. ...