Inspiration
Everywhere we look, people struggle with unhealthy eating—fast food, skipped meals, and processed snacks. It’s easy, it’s convenient, and it’s a cycle that’s hard to break. But how do you break an addiction? With another addiction. Nomzi turns healthy eating into a game. Scan your meals, compete with friends, and collect badges for building better habits. It’s not about dieting—it’s about making smart choices fun. Even just one healthy meal a day makes a difference, because small changes add up, and the right choices today will shape a better future.
What it does
Nomzi makes meal tracking fun, competitive, and rewarding—helping you build better habits while challenging friends. With Nomzi, users log their meals, earn badges, and reach milestones for making healthier choices. Every meal tracked brings them closer to new achievements, keeping motivation high. The app adds a social twist—friends can join challenges, compare progress, and push each other to stay on track. The more consistent you are, the more badges you unlock, turning smart choices into a winning streak. Because healthy habits aren’t built overnight—but with the right motivation, they stick for life.
How we built it
Nomzi is a full-stack app we built by integrating AI-powered food recognition and user-friendly mobile app design. Our development process started with implementing Groq's API to leverage the computational power of LLaMA in the backend using Python, which serves as the core engine for food detection and personalized meal recommendations. For the frontend, we designed the app in Figma and converted it into React Native using a plugin. The app runs on Expo Go, enabling real-time testing and smooth cross-platform performance. To manage user data, we utilize MongoDB, which stores both food images and meal-related information captured by users. This ensures scalability and quick retrieval of data, enhancing the overall user experience.
Challenges we ran into
During backend development, we encountered challenges integrating Generative AI into our final project. Implementing Meta's LLaMA API required thorough testing to determine which prompt structures yielded the best and most reliable responses for food recognition and meal recommendations. Fine-tuning the AI model to provide accurate and relevant results was a crucial part of this process. On the frontend side, converting Figma designs to React Native proved to be far from seamless. Even after using a conversion plugin, many UI elements and effects did not translate properly, leading to unexpected bugs. We had to manually inspect and modify multiple files, adjusting styles, layouts, and animations to match our original design vision. This required debugging React Native components and ensuring that all UI interactions were properly implemented.
Accomplishments that we're proud of
We are incredibly proud of the effort we put into learning and implementing software tools that were completely new to us in order to bring Nomzi to life. At the start, the process was challenging and unfamiliar, from integrating Generative AI in our backend to converting Figma designs into React Native for our frontend. Never have any of us built a mobile app before, but through perseverance and dedication, we were able to deliver a final product that we are all proud of.
What we learned
One of the biggest lessons we learned from this experience was the importance of better time management when working with unfamiliar technologies. We underestimated how much time it would take to debug the Figma-to-React Native conversion, which nearly prevented us from using the UI we had carefully designed. Moving forward, we know that when tackling new tools, we need to set aside extra time for troubleshooting and iteration, ensuring we have the flexibility to adapt and refine our work without last-minute stress.
What's next for Nomzi
We truly believe Nomzi can change the way people think about health and nutrition. By using AI to make healthy eating more intuitive, we hope to help users make better food choices effortlessly. Moving forward, we plan to expand our food database, allowing for more accurate meal recognition and smarter recommendations. We also want to let users share their own meals and recipes, creating a community-driven space for healthy eating. In the future, we’re excited to explore integrations with health apps, making Nomzi an even more powerful tool for tracking nutrition and wellness. This is just the beginning, and we can't wait to see how Nomzi grows!
Built With
- expo-go
- figma
- git
- github
- groq
- llama
- mongoldb
- python
- react-native
Log in or sign up for Devpost to join the conversation.