Welcome!
Hello and Welcome! I started in applied AI six years ago when we did almost everything by hand, talk counting tokens. A lot has changed.
What I write about is the gap between AI theory and what actually happens when you bring it into a real business: small, medium, with real data, real constraints, and a vague idea about AI. The failures are real and the fixes are hard-won.
This blog is my way of working through those experiences: part field notes, part reflection - for anyone curious about what building with AI actually feels like on the inside.
Recent Writing
View all →The Problem With Labels Is Everything Before the Labels
Four problems stacked on top of each other. Label design, subjectivity, resolution, and redundancy. What survived was 16 labels - and a clearer understanding of what unstructured data can actually add.
2026-03-04315,000 Properties Lost
Three bugs stacked on top of each other. Each one only visible after the previous was fixed. Here is what happened and what bulletproof checkpointing actually looks like.
2026-03-03What Happens When You Run an LLM on 1.5 Million Texts
Chinese hallucinations, 20+ artifact patterns, and the six-layer fallback system I built to get 1.4M clean translations.
2026-03-03