I Built an AI Agent that Searches Thousands of Autonomous Driving Videos in 2 Seconds
Get the Code: github.com/mongodb-industry-solutions/manu-multimodal-event-explorer The Multimodal Event Explorer uses AI and unified search to find 1-in-a-million autonomous driving scenarios (edge cases) in seconds, replacing manual filtering that used to take weeks. 1 Built on MongoDB Atlas for vector, full-text search, and metadata filtering, this application demonstrates a conversational AI agent that can execute complex search and comparison queries in plain English. 1 Watch the video for a full live demo and a breakdown of the architecture, data model, and hybrid search pipeline. The full source code and setup instructions are open source and available on GitHub. 00:00:00 - Introduction & Tech Stack Overview 00:00:41 - The Challenge of Autonomous Driving Edge Cases 00:01:25 - System Architecture & The 3 Layers 00:02:21 - MongoDB Atlas Data Model & Document Structure 00:03:57 - Vector Compression via Scalar Quantization 00:04:17 - Live Demo: Hybrid Search & Filter Stacking 00:05:43 - Live Demo: Conversational AI Agent & Safety Workflows Subscribe to the MongoDB for Developers YouTube Channel: https://www.youtube.com/@MongoDBDevelopers?sub_confirmation=1 Sign-up for a free cluster → https://www.mongodb.com/cloud/atlas/register Subscribe to MongoDB YouTube→ https://mdb.link/subscribe Visit Mongodb.com → https://mdb.link/MongoDB Read the MongoDB Blog → https://mdb.link/Blog Read the Developer Blog → https://mdb.link/developerblog MongoDB for Developers YouTube Channel → https://www.youtube.com/@MongoDBDevelopers