Five software engineering roles for working with AI

Four years into the AI boom, the way software engineers interact with AI is still evolving. We’ve moved from better code completion to autonomous agents that triage incoming bug reports, determine root causes, and deploy fixes. Along the way, developers have divided over how much interaction is required to produce high-quality outcomes. Some prefer a direct, hands-on approach, while others prefer to stay hands-off. In reality, there’s a wide spectrum of possible interaction models between humans and AI. I like to think of these models as the familiar collaboration roles I’ve experienced in my career. Observer (pair programmer) Human pair programmers traditionally sit next to each other at a computer. One person is the driver, who writes the code, and the other is the observer, who reviews it in real time. These days, pair programming can also happen remotely through screen sharing. When working with AI, you can take the observer role and actively watch as it writes code. You can stop…