In T308164, the ML team hosted the recommendation-api which was built using the Flask web application framework. In T369484, the Language team rebuilt the recommendation-api using the FastAPI web application framework. We are going to deploy this modernized API on LiftWing in two phases:
- Staging: The Language team will test the endpoints, identify and report any encountered edge cases, and collaborate with the ML team to resolve these issues to ensure the API is ready for production use.
- Production: After successfully completing the staging phase, we will deploy it to production for full-scale use through the WMF API Gateway.