Writing
Thoughts on AI, engineering leadership, and building products.
-
The 48-Hour Laboratory: Why Hackathons Are the Missing Operating System for AI Experimentation
AI adoption does not fail because organisations lack intelligence. It fails because our learning loops are too slow, our fear of being wrong is too high, and our teams rarely get a protected space to try, fail, and learn together.
-
Respecting the Craft: How to Convince Software Engineers to Use AI
The hardest part of AI adoption in software engineering is no longer the models. It is persuading proud, experienced engineers to let those models into their daily craft without feeling slower, deskilled, or disrespected.
-
Scaling AI Engineering Teams
Building high-performing AI teams requires more than technical expertise. It demands intentional culture-building, clear ownership models, and the ability to balance research exploration with production delivery.
-
Leading Distributed AI Teams
Remote AI teams need different communication patterns than traditional engineering orgs. Async-first documentation, clear experiment tracking, and outcome-focused cultures become critical.
More posts coming soon...