Completed

Movie Recommender

End-to-end MLOps pipeline for movie recommendations.

Period: Sep – Nov 2025
Role: Final project for the MLOps certificate
Status: Completed
Stack: Python
TruncatedSVD
FastAPI
Streamlit
MLflow
Airflow
MySQL
Docker

Streamlit interface with movie recommendations

Overview

A recommender system from data to delivery: the model (TruncatedSVD) is trained, versioned and served through an API; an interface makes the recommendations visible.

Building blocks

Model: matrix factorisation with TruncatedSVD.
Delivery: FastAPI as the API, Streamlit as the interface.
MLOps: experiments in MLflow, pipelines with Airflow, data in MySQL, all with Docker.

Details on data, metrics and an architecture diagram will follow.