My AI Adoption Journey

Mitchell Hashimoto walks through his own adoption in six named phases: drop the chatbot, reproduce your own work, end-of-day agents, outsource the slam dunks, engineer the harness, and always have an agent running. The framing is that each phase was uncomfortable before it was useful, and that the early ones felt like a cost rather than a gain. On reproducing work he already knew how to do:

This was excruciating, because it got in the way of simply getting things done.

He is careful not to oversell the later stages. Background agents running continuously are, by his estimate, effective 10 to 20% of the time, which is worth it only because the attempts are cheap and run while he does something else. He also budgets a fixed daily slot, around 30 minutes, for setting up end-of-day runs.

The practical notes are the useful part. Turn off agent desktop notifications, since context switching is expensive. Build the harness, meaning docs and tooling, so agents stop making the same mistakes. And keep hand-writing the code you actually want to write: he describes himself as a software craftsman that just wants to build stuff for the love of the game.

He declines to settle the question of whether leaning on agents erodes skill, saying the pace of model progress forces him to keep revisiting his priors on it. Pairs well with Boris Cherny’s shorter steps of AI adoption, which maps the same terrain as four organisational steps rather than one person’s arc.