MacroMaster

MacroMaster lets you snap a pic of your meal to instantly track calories, carbs, protein, and fat. Set goals, view logs, and stay on top of your nutrition—all in one smart, easy-to-use web app.

Comment 1

Inspiration

We love going to the gym and seeing progress, but we realized diet plays just as big of a role as exercise. Tracking calories helps stay on track to your goals, so we decided to make a tracking app that's quick and easy to use.

What it does

MacroMaster helps users track their daily calorie intake using food image uploads. It detects the food items and prompts users for food volume. It calculates calories and macros (carbs, protein, and fat), and helps users set and maintain long-term nutrition goals. The app also stores and visualizes past logs so users can stay consistent.

How we built it

We built the frontend using React, developed the backend using Flask, and stored user data with a PostgreSQL database. For the AI side, we used Ultralytics’ YOLOv8 for food object detection and integrated the model's output with USDA FoodData Central to retrieve accurate nutritional information. Everything was connected via REST APIs, enabling seamless data flow between the client, server, and model.

Challenges we ran into

Along the way, we ran into a few challenges. One of the toughest was finding a good dataset for food object detection that would work well with YOLO. We also had to iron out a few version control issues as we worked collaboratively on GitHub, especially when managing branches and handling pull requests under time pressure.

Accomplishments that we're proud of

We built a fully functional app that combines AI, web development, and real-world nutritional data. Additionally, we successfully integrated our custom detection model with USDA’s food database, created a clean and user-friendly interface, and developed a robust backend to support login systems and data storage.

What we learned

Throughout the process, we learned a lot—especially about managing collaborative coding projects. We improved our use of Git and pull requests, developed strategies for handling time crunches, and deepened our understanding of YOLO and object detection in applied machine learning.

What's next for MacroMaster

Looking ahead, we plan to polish and deploy MacroMaster to make it available for real users. Our goal is to turn this into a full-scale product, monetizing through ads to keep it free for everyone. We believe MacroMaster can make a real difference in helping people eat smarter, live healthier, and achieve their goals.

Share this project:

Updates