FixIt: AI Troubleshooter for Non-Technical Friends
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built I built HomeWatch , a local-first AI assistant that helps my friend keep an eye on their home and, more specifically, their dog while they're away. The problem was simple: they have a camera pointed at the main living area, but checking through hours of footage to figure out what happened during…
This submission for the Hacktoberfest Open-Source AI Challenge Week 1 showcases HomeWatch, an AI assistant designed to help users monitor their homes and pets without the need to manually review lengthy camera footage. Created by a user named tba, HomeWatch processes video locally and turns it into a simple timeline of events, enabling users to ask questions such as "What did the dog do while I was out?" or "Was there anything unusual?"
The system can detect and summarize relevant events, including unusual behavior like the dog spending an unusually long time by the front door. Importantly, the camera footage remains within the user's home and is not sent to a third-party AI service, ensuring privacy and offline operation. HomeWatch is built using open-weight AI models running locally and consists of a camera stream, video stream, frame/event detection, person/animal/object/object detection, scene changes, event extraction, local vision model, event database, and a local language model coupled with a retrieval-augmented generation (RAG) system.
The camera stream is processed locally, and relevant events are extracted without sending the entire video history to an external service. Events are stored with timestamps and contextual information, allowing the language model to retrieve relevant events when queried. For instance, a question like "What did Luna do between 2pm and 4pm?" can be answered using the locally generated event history instead of analyzing two hours of video each time.
The AI model can be replaced based on the user's hardware capabilities, from a powerful desktop to a dedicated home server. The open innovation approach of HomeWatch offers several benefits, such as ensuring private data remains within the user's home, enabling offline operation once models are installed, eliminating per-video API costs, allowing users to experiment with different models, and providing control over the system's detection, retrieval, prompting, and model layers.
The project was developed using an AI coding agent to assist in designing the event-processing pipeline, implementing local inference workflow, and building the natural-language query interface.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
