Skip to content
Blog

Train Different.

Thinking out loud about ML engineering, provenance, and what it means to build reliable systems in a world that runs on learned behavior.

Market Structure · Open Models
NVIDIA + Hugging Face: Who Owns the AI Build Workflow?

NVIDIA reportedly agreed to buy the place open models live. That settles one question about the AI stack and leaves a larger one open — where teams actually propose, review and reproduce a build.

Aug 27, 2026 · 6 min read Read →
Benchmarks · Provenance
Tracing a Training Run Costs 1%. The PCIe H100 "Discount" Costs 89%.

Purify and Valgrind made tracing infamous. We measured roar on real H100 pretraining: about 1% on a three-hour run. The expensive variable turned out to be the one nobody had put a number on.

Jul 14, 2026 · 6 min read Read →
Provenance · Source of Truth
Can We Rebuild ML's Source of Truth?

Git gave software engineering a record nobody has to maintain. ML never got one — and the reason isn't that the problem is harder. Two primitives recover what Git gave us.

Jul 2, 2026 · 6 min read Read →
Provenance · Source of Truth
Git Solved Provenance. AI Quietly Broke It.

Git made provenance objective and automatic — history as a byproduct of doing the work. Then the models crept in. Every MLOps tool since is a crutch that rebuilds provenance before or after the run. The fix is during.

Jun 25, 2026 · 5 min read Read →
Provenance · ML Engineering
How Provenance Quietly Erodes in ML Engineering

Every ML system is a chain — data, pre-processing, extraction, model. Trust lives or dies at the seams. Three approaches to provenance today, and why all of them leak.

Feb 26, 2026 · 8 min read Read →