I'm a Swedish engineer based in Ho Chi Minh City. I spent a decade shipping production software to a million users as Head of Engineering, leading the team that built and scaled it. That experience taught me more about where systems fail than any course or book ever did.
This blog is where I document what I'm building, the decisions behind it, and what breaks along the way. Right now that means an autonomous engineering pipeline: a system where specialist agents take a ticket from input to tested, reviewed code with no human in the loop, proven on real production codebases, including a ~100k line TypeScript monorepo. Safely modifying any part of a codebase that size requires reasoning about all of it. The design decisions come from having been burned by the alternative in production. Hard stops, agent isolation, TDD enforcement: none of those are academic choices.
The posts focus on the reasoning, not the internals. Code appears where it earns its place.