AI-powered lost and found matching platform for campuses and organizations
LostLink is a full-stack web application that uses semantic vector search and machine learning to intelligently match lost items with found items. The system combines PostgreSQL with pgvector for embedding storage, sentence-transformers for text analysis, and Google's Generative AI for contextual reasoning, providing ranked match suggestions with explainable scoring.
When students lose their wallets, phones, or backpacks on campus, they have no easy way to check if someone turned them in. Campus security or lost-and-found offices keep physical bins with items, but matching a found item to its owner requires manually digging through descriptions that use different words (someone might write 'dark school bag' while the owner reported a 'black Nike backpack'). Without a smart system to connect these asymmetric descriptions, items sit unclaimed for months and owners never find out their belongings were actually recovered.
LostLink addresses the challenge of traditional lost-and-found processes by integrating machine learning and AI to intelligently match lost items with found items. By using semantic vector search to understand various descriptions and leveraging contextual AI for match explanation, the platform creates a more effective and user-friendly system that significantly increases the chances of items being returned to their owners.
This isn't just a project — it's a simulation of how software actually gets built inside a company. You'll see how real teams gather a client's requirements, negotiate trade-offs, and ship in sprints instead of all at once, while picking up the daily rhythm of an engineer's (and tester's) week along the way.
One purchase, the entire toolkit — nothing held back, nothing sold separately.
6 architecture & design diagrams(HLD, LLD & more) — unlocked with the Standard or Premium plan.
No demo video yet
The creator hasn't uploaded a walkthrough video.
