Hermes Movie Recommender

A full-stack application that provides general movie suggestions based on content similarity.

Overview

This project is a high-performance movie recommendation engine. It leverages AI-powered text embeddings and vector similarity search to find the most relevant films based on semantic meaning rather than just shared genres or titles.

Tech Stack

• Frontend: React 19, TypeScript, and Vite.
• Backend: Python 3.12.
• Database: PostgreSQL 18 with the pgvector extension.

Key Features

• Semantic Search: Uses text embeddings to understand the context of movie descriptions.
• Vectorized Ranking: Utilizes pgvector to perform fast nearest-neighbor searches across candidate features.