Inspiration
We were inspired by a simple problem: small businesses often realize they are running low on important materials only after an upcoming order creates a shortage. We wanted to build an AI-powered prototype that could look ahead and help prevent that problem.
What it does
StockSense AI connects current inventory with confirmed upcoming orders and supplier information. It forecasts material demand, identifies potential shortages, calculates reorder needs, compares supplier options, and prepares a clear recommendation for the business owner.
The final purchase decision always stays with the human, who can approve, modify, or ignore the recommendation.
How we built it
We created StockSense AI around a small bakery scenario. The prototype uses mock inventory data, upcoming orders, bill-of-material requirements, supplier information, and simple business rules.
The interface was built as a web prototype using Google AI Studio and TypeScript. The project code is hosted publicly on GitHub.
Challenges
Our main challenge was making the project realistic while keeping it simple enough to build and demonstrate within the hackathon timeframe. We had to make sure inventory levels, upcoming demand, reorder quantities, and supplier options worked together in a clear decision flow.
We also wanted the system to avoid making unsupported assumptions when information was missing.
What we learned
We learned that an AI inventory agent needs more than a conversational interface. It needs structured data, clear rules, reliable calculations, and defined decision boundaries.
We also learned the importance of keeping a human involved when a recommendation can lead to spending money.
Result
The result is a practical hackathon prototype that demonstrates how StockSense AI can move from detecting a potential inventory problem to preparing a clear replenishment recommendation for a small-business owner.
Log in or sign up for Devpost to join the conversation.