What's next for Protein+

The 137g Protein Architect is a high-performance AI nutrition engine built for the Google Gemini Hackathon at AITR. Designed specifically for Indian vegetarians, it solves the challenge of hitting a strict 137g daily protein target while navigating a local diet. Powered by Gemini 1.5 Flash, the app uses multimodal vision to instantly identify regional dishes—from Indori Poha to Paneer—providing a high-precision protein breakdown from a single photo.

Beyond simple tracking, we introduced the Protein Efficiency Score (PES). Borrowing from priority-scoring logic, this feature grades each meal on a scale of 1-10 based on its protein density. If a meal falls short, the "Architect" proactively recommends Antigravity Boosts—localized suggestions like soya chunks or Greek yogurt—to bridge the gap before the day ends. Built with Streamlit for a seamless UI, our project transforms standard meal logging into a data-driven performance roadmap. By leveraging Gemini's low-latency inference, we've created a tool that understands the specific nutritional nuances of the Indian student community, ensuring every bite counts toward their fitness goals.

Built With

Share this project:

Updates