EchoMind – AI Visual Memory Assistant

We've all experienced the frustration of misplacing everyday items — keys, wallets, glasses — and wasting time searching for them. Existing reminder apps can tell you what to do, but none of them remember where physical objects actually are. EchoMind solves this by acting as a visual memory for your space. Using YOLO based real time object detection, it continuously observes your environment and logs when and where objects were last seen. A conversational AI layer, powered by Qwen and Llama models via the Fireworks AI API, then lets you simply ask, "Where did I leave my keys?" and get an instant, natural-language answer — complete with greeting and small-talk handling for a smooth, human like interaction. Built with a Next.js frontend (deployed on Vercel) and a FastAPI backend (deployed on Render), EchoMind runs on AMD Developer Cloud with ROCm, combining computer vision and language models into a single, practical memory augmentation tool. Built by a two person team for the AMD Developer Hackathon: Act II — Unicorn Track, EchoMind shows how pairing object detection with a Fireworks-powered conversational agent can solve a small but universal daily frustration.