AI Native Engineering
The patterns, maturity model and leadership strategies behind Horizon's AI for Productivity programme - and what we got wrong along the way.
In May 2025, a small group of engineers inside Horizon, part of Meta’s Reality Labs, asked a simple question: how can AI actually help us hit our engineering excellence goals? Seven months later we had a 400-person community, a maturity framework, and hard data on what worked - 80% time savings on specific workflows, 81% weekly tool adoption, and replicable patterns for test coverage, large-scale refactoring, custom context integration and autonomous code fixes.
That was December. By the time I gave this talk the models had got better, I’d moved to a different part of Meta, and some of the things I was confident about had aged badly. Others held up well.
I shared the patterns that reliably delivered results, the maturity model we used to help teams assess and grow, and the leadership strategies that turned scattered experimentation into systematic adoption. I also talked about what we got wrong: the trust and quality problems that dominated workshop discussions, the code review bottleneck that AI-generated diffs created, and why we threw out our original metrics.
The Hungarian Railway Museum is quite a venue for a software conference. Thank you to the Craft team for having me.