NutriGuide | Devpost

Inspiration

In today’s day and age, processed foods have become more popular than ever and negative health conditions are proportionally rising. According to the Food Research and Action Center, the latest data indicate that 39.6 percent of U.S. adults are obese. Many people face the problem of deciding between their dietary wants and needs. In order to facilitate this decision, we decided to create NutriGuide, an app that adapts to both the user’s food preferences and needs to create a suitable meal plan.

What it does

NutriGuide takes in user input such as weight, height, and dietary restrictions to provide users with a personalized meal plan. Our app recommends the healthiest food that aligns with the user’s recommended daily calorie intake based on their input. After a recommendation is generated, the user can change the meal plan according to their preferences e.g. less carbohydrate consumption or more protein intake.

How we built it

NutriGuide was built using Python and OpenAI. On the backend, the OpenAI API was used to utilize ChatGPT 3.5 to process the user’s requests. It took in the parameters of the app such as user height, weight, age, gender, and dietary restrictions to generate a plan. On the frontend, the platform was built using tkinter and custom tkinter. These modules allowed us to work on creating a simple interface that could take in user data and allowed the user to give feedback to improve results.

Challenges we ran into

Frontend-We didn’t know how to create a GUI so we had to learn tkinter and custom tkinter on the go which was time consuming, but helpful to develop the frontend application. Backend-We are not very familiar with the interaction with chatGPT. However, we did the research and incorporated this into our project because we want to include this powerful tool to enhance the result.

Accomplishments that we're proud of

Learning about the frontend and backend aspects of a project and developing consistent results was an exhilarating experience that significantly expanded our skill set. This was our first hackathon, and we are proud to have successfully completed our project, showcasing our ability to collaborate effectively, think creatively, and deliver a solution that meets our goals.

What we learned

We gained proficiency in integrating OpenAI’s GPT strengths with various APIs to streamline output generation within our project. Additionally, this marked our initial exposure to frontend development. One team member successfully acquired skills in using tkinter and creating custom tkinter elements in Python to design the user interface all during the span of the competition.

What's next for NutriGuide

In the future, we plan to increase the number of dietary restriction options to raise accessibility and provide detailed feedback on nutrient consumption by breaking it down into macronutrients. We would also like the app to provide the user with information about the carbon footprint generated by each meal that is recommended.