A Large Short-video Recommendation Dataset with Raw Text/Audio/Image/Videos (Talk Invited by DeepMind).

Recommender system finds its application in many aspects of the online ecosystem including product recommendation, movie recommendation, books, news, video recommendation to name a few.

The YouTube Video Recommendation App is built with React and Vite. This application fetches and displays a list of recommended videos based on a given category ID using the YouTube Data API.

A FastAPI-based video recommendation system that connects to the EmpowerVerse API to provide personalized video recommendations.

A YouTube browser extension for navigation and analysis of video content & comments using Python, Node.js, HTML5 & CSS.

Video Recommendation Engine is an advanced AI-powered system designed to deliver highly personalized video content based on user preferences, engagement history, and emotional state. Inspired by the Empowerverse App, this engine leverages Graph Neural Networks (GNNs), deep learning, and content-aware filtering to serve motivating and mood-aligned.