We were inspired by the health vertical and Vision Recognition technologies. We wanted to create a very simple and user friendly way to track calorie and nutrient intakes, hoping to make it as easy as possible to develop and sustain healthy eating habits.

About Our Product

Using Clarifai's food model vision recognition, NutriBot's mobile app detects the types of food present in a picture the users take on their phones. It then prompts the user with a list of predictions of the contents of the image and asks the user to choose the correct ones among the list. Then, using Nutritionix API, it gives a breakdown of the calories and nutrition values for each food item. On a separate web app, the historical data is displayed to the users to inform them about their long term trends in food consumption.

How We Built It

The mobile app was built using React Native, Clarifai Food Model API, Nutritionix API, and Expo. The web app was built using Wix Code.


Finding the right platform took multiple trial-and-error. We explored everything from Flask to Wix Code and graphing applications like Charts Ninja and chart.bundle.js. As most of us were unfamiliar with these new platforms and JavaScript, our biggest challenge was learning those skills and being able to apply them in a timely, scalable manner.


Although our app still has many more features to come, it well encompasses our initial vision with our 36 hours of hacking. We took every hour as an opportunity to learn by struggling, exploring, and collaborating.

What We Learned

Every expert starts out as a beginner! Coming in, we didn't know everything we needed to know for our project. But through persistence and proactiveness, we were able to turn the gaps in our knowledge into a motivating source for learning. From this weekend, we are taking away Wix Code, JavaScript, Flask, HTML, and React Native, which we hope to then apply to our future hacking endeavors.

What's Next for NutriBot

We aim to be able to give users a more thorough analysis of their food consumption trends such as their daily calorie intakes, calorie values broken down by specific nutrition values like protein, carbohydrate, fats on weekly and monthly bases. We also hope to be able to give users instant feedback about their meal plates, for example if the user's plate only has fat-heavy food, we want to display a warning that suggests them to consume more vegetables and/or protein.

Share this project: