Fine-tune Qwen3-4B on your own posts with MLX on a Mac: data prep, the LoRA config I used, a blind test, and where a model like this still fails.
Badges
Badges come from GitHub once the author claims this feed.
Recent posts
every link opens the author’s post-
How to Fine-Tune a Local LLM to Write Like You (on a Mac)
-
Does Your Fine-Tuned LLM Sound Like You? How to Test It
How to test whether a fine-tuned LLM writes like you, with blind LLM judges.
-
PHP vs JavaScript: which is the worst programming language?
Two prosecution briefs, 69 arguments, every code sample executed or sourced, then a rebuttal round where each side defends itself with the other's ammunition.
-
Where to put your WordPress database password
WordPress hands the database password to MySQL on every request, so it has to be readable by PHP - which is why every hardening tip just moves it somewhere else.
-
A shoutout to Stario, the Python framework behind 12 of my projects
Twelve of my projects run on the same Python framework - Stario, written by my friend Adam Bobowski.