Category recommender v2
Replacing a FastText classifier with LLM classification
Context
The feature suggests a category when a seller lists a product — successor to the v1 I built as an intern.
Problem
The FastText-based recommender had hit its ceiling on both accuracy and extensibility.
How I solved it
- More data and another training run would have fixed accuracy alone, but the deeper constraint was a taxonomy that keeps changing
- So I removed training from the loop entirely, replacing the FastText classifier with LLM-based classification
Result
- Adding a new category no longer requires retraining, so the taxonomy can grow freely
Impact
The taxonomy can change without anyone retraining a model.