Major project
Retail, E-commerce & Consumer Goods
Product recommendation service in production
Build a recommender and deploy it as an API with caching, tests and click-through monitoring.
The brief
A fashion retailer wants personalised recommendations on every page, served in milliseconds. Build a recommender from purchase history, then deploy it as an API with caching, tests, versioned models and monitoring of click-through. Good looks like: offline evaluation with precision@k, a fast API, a canary release, and a click-through dashboard.
What you will build
- Build candidate retrieval and ranking from purchase history
- Evaluate offline with precision@k and coverage
- Serve recommendations from a cached API
- Add tests and register model versions
- Release a new model as a canary
- Monitor click-through and latency
Free, open data to build it with
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.