Inspiration
When thinking about what is important to students the first thing that comes to mind is health. However, as full-time students, health often falls off our priority lists. Our teammate, Max Chwa, told us about how students can feel overwhelmed by the amount of Dining Halls and available meal plans, that considering what is the healthiest option for them and for their health goals can feel impossible. NutriLion was meant as an aid in achieving this goal.
What it does
LionDiet is a website intended to help Columbia students with access to a meal plan decide what their next meal should be based on their nutritional goals. They will provide the site with their age, weight, height, weekly exercise amount, and intended health goal. LionDiet will recommend an available Dining Hall and meal. LionDiet operates by scraping LionDine, a daily calendar and menu of all available meals at the Columbia Dining Halls. It will then call groq.api to obtain the nutritional value of all the meals. Here are the nutritional facts assessed by groq: Protein Carbohydrates Fats Fiber Sugars Sodium Saturated vs. unsaturated fats Micronutrients LionDiet will take the user's input and use groq to assess the nutritional goals of the user, how much protein should they consume? How much carbohydrates? And so forth. After having considered the optimal diet for the user, LionDiet will use groq to compare the needs of the user with the available meals in the Dining Halls and give back a recommended meal, or meal plan for the day.
How we built it
We scraped LionDine.com using a web scraping library in Python. Using that, we used groq’s api to give us nutritional data on the meals. We used StreamLit to build the frontend, and had groq’s api, to consider what would be the ideal diet for the user. We then used groq’s api to compare the ideal diet with the nutritional values of the available meals to recommend the best meal.
Challenges we ran into
A majority of our team has limited coding experience. Most of us have very limited exposure to Python. As such, it was difficult to program everything in time.
Accomplishments that we're proud of
None of our group members have attended a Hackathon prior to DevFest '25. Coming up with an idea for our project and executing it within such a short span of time has been a tremendously satisfying process.
What we learned
We gained a better understanding of Gen AI and the complexities involved in finding a prompt that can achieve the results we desire.
What's next for NutriLion
Thinking about dietary restrictions (halal, kosher....) Create an account to store your previous meals to generate a more personalized experience.
Log in or sign up for Devpost to join the conversation.