Campus Shield

CampusShield is a web app designed to improve campus safety by detecting potential weapons in video footage and turning those detections into alerts that can be reviewed in a dashboard. I was inspired by the idea of using AI and software to help create faster situational awareness in environments where safety matters.

I built the project by combining a machine learning detection pipeline with a web-based frontend so detections could be surfaced in a more practical and usable way. Through this project, I learned more about integrating ML models into an actual application, handling large datasets and model files, and connecting backend outputs to a clean user interface.

One of the biggest challenges was managing model performance and project size at the same time. Training and testing the detection system required large files and generated outputs, so I had to think carefully about performance, organization, and what should or should not be included in deployment and version control.

Built With

• api
• ml
• python
• supabase
• typescript