GluCUE

Inspiration

In our team, several of us have loved ones who have been diagnosed with diabetes. We have seen their struggles firsthand, they wear glucose sensors that show their sugar levels, but these devices don’t tell them what to eat or how to manage their diet based on their glucose level they have. This leaves them without clear guidance on what foods will help them manage their glucose levels. Watching them face these daily challenged inspired us to create GluCUE, AI powered tool that turns glucose data into personalized meal recommendations and nutrition insights.

What it does

Glucue is a web application that helps people with diabetes and prediabetes turn continuous glucose monitor (CGM) data into personalized dietary guidance. Users securely connect their Dexcom or Libre device, complete a quick dietary survey about their eating habits, and set personal glucose goals. Glucue continuously pulls real-time glucose readings, analyzes them alongside logged meals, and displays trends through a clean, mobile-optimized dashboard. Over a few weeks, the app identifies foods and meal timings that trigger spikes or help maintain stable glucose levels, generating a fully personalized dietary plan.

If users eat outside of the plan, they can simply snap a photo of their meal, and Glucue uses Gemini AI to predict the expected blood sugar impact instantly, analyzing the meal’s likely effect based on ingredients and past glucose responses. Gemini also assists in creating smarter meal recommendations and trend analysis, helping the app adapt to each user's body faster over time.

Built with React.js, Tailwind CSS, Node.js, and integrated with Dexcom’s API and Gemini AI, Glucue ensures a fast, intelligent, and scalable user experience. Future plans include a clinician dashboard for remote patient monitoring and partnerships with healthcare providers to reward users for maintaining better glucose control through insurance incentives.

How we built it

We built GluCUE by connecting glucose sensor data to backend that generates AI powered personalized meal plans. To analyze food photos, calories, nutrition, and predict glucose spikes, we used Gemini 2.0 Flash (Google’s multimodal model) to power the analysis pipeline. Users can upload a meal photo, and the AI instantly detects the food, estimates its nutritional content, and predicts its impact on blood sugar levels.

Challenges we ran into

One major challenge was connecting the frontend and backend to work smoothly. Since we used a Dexcom glucose dataset to simulate real-time sensor data, and real-world uploaded meal images for nutrition analysis, we had to carefully design the system so that both data sources worked together without errors or delays.

Another challenge was ensuring that the AI-generated meal recommendations were accurate and personalized based on the user's current glucose levels. We also needed to make the experience easy to understand for users without overwhelming them with too much technical or medical information.

Accomplishments that we're proud of

We successfully brought a niche and impactful idea to life by combining real-world food image analysis with glucose monitoring insights for diabetic users. Unlike typical sensor apps that only report glucose numbers, our project aims to analyze the food itself and predict glucose risk, helping users make healthier decisions before eating.

What we learned

This was the first hackathon for all of us, and none of us had experience hosting a website before. We had to learn how to set up and deploy a web application as we built it, facing real technical challenges along the way. We also gained hands-on experience with frontend development — from structuring pages in React to designing responsive layouts with Tailwind CSS. Beyond just technical skills, we learned how to collaborate under tight deadlines, adapt quickly, and turn a rough idea into a working solution we’re proud of.

What's next for GluCUE

Moving forward, we aim to grow our platform into a full-scale health companion for anyone using a glucose sensor. In addition to meal planning and nutritional guidance, we plan to introduce features that help users track changes in weight, manage medications, monitor physical activity, and keep an eye on key health indicators like kidney and eye health. We also want to expand the emotional connection users have with their companion by introducing customization options such as unlockable accessories and evolutions, rewarding healthy habits and personal milestones. Our ultimate goal is to create a platform that not only provides real-time guidance but also inspires and empowers users to take control of their long-term health in a rewarding and engaging way.