The full methodology behind localscrub's stage-2 specialist: synthetic training data with exact...
The full methodology behind localscrub's stage-2 specialist: synthetic
training data with exact labels, training on the serving distribution, and
why the fine-tune's first product is not accuracy - it's parseability.
In the last article I benchmarked localscrub, a local-first PHI de-identification cascade, and buried a teaser near the end:
a 1.7-billion-parameter model, LoRA-tuned in 17 minutes on a consumer
Part 1 of a 4-part series. Full fine-tuning a tiny Gemma 3 model for intent classification — the generative framing, the…
Parsed fine-tuned a 27B open-source model to beat Claude Sonnet 4 by 60% on a real-world healthcare task—while running 10–100x…
Your data never leaves the machine - and you can check my math Building localscrub, a...
Part 4 (finale) of a 4-part series. Three model sizes tied on the same task — so when does bigger actually earn its keep? And the…
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Part 2 of a 4-part series. How LoRA works (the low-rank trick), a working PEFT config, and three real GPU walls I hit — the FP16…
