All projects

Automating LLM translation cleanup for product data

Cleaning up product translations inside Windly

ContractAbearApr 2024 – Jun 20243 mos

Context

Windly surfaces overseas products to domestic buyers, so translation quality is product-page quality.

Option names, titles, and tags were each translated their own way, so the same product went by different words on different screens.

How I solved it

  1. Switched the approach from rule-based to LLM-based cleanup, designing and building the pipeline on Langchain
  2. Proved it internally with a Streamlit MVP first, then built the serving backend

Result

  • The cleanup pipeline shipped to production inside Windly
  • Applied across roughly one million products
    • option names, titles, and tags

Tech

PythonLangchainStreamlitFastAPI