BoilerPlate

Main Page
Sign In Page
Sign Up Page
Generation Page
Settings Update Page

Inspiration

Purdue dining courts offer an abundance of variety food options. Sometimes it can be overwhelming to have the best, healthy, and balanced meal based on your dietary and physical needs. To simplify this, we built BoilerPlate.

What it does

We leverage the Purdue dining court API in conjunction with OpenAI's Chat Completion API to create personalized recommendations for the healthiest and most balanced food options available at your selected dining court. Utilizing your age, weight, height, gender, and weekly exercise frequency, we tailor our suggestions to meet your specific preferences and nutritional needs.

How we built it

We used React.js, integrated with Next.js, to develop a comprehensive full-stack web application. Our backend is powered by Firebase and the OpenAI API, while we seamlessly gather detailed information about available foods from Purdue's dining court API. The construction of our UI was facilitated through the extensive use of Tailwind CSS.

Challenges we ran into

Since this was our first time using OpenAI's API, we struggled to generate a consistent output that required actual prompt engineering. Additionally, we encountered specific issues related to React that we prefer not to delve into further.

Accomplishments that we're proud of

Developing our first project that incorporates artificial intelligence.

What we learned

We learned a lot by using new tech stacks including TypeScript and Next.js. Also, we learned how to efficiently divide our parts into a team of four people, leading to an agile development process. Not to mention, utilizing OpenAI's API and Purdue dining court's API to obtain the desired output was what we learned most technically.

What's next for BoilerPlate

We have plans to expand our feature set to encompass off-campus restaurants near Purdue and, if feasible, extend our reach to include other college campuses as well.

Built With

chatgpt
firebase
html
next.js
openai
react.js
tailwindcss
vercel

Submitted to

Boilermake XI

Created by

Prior to starting the project all of us discussed the details on how we approach the project. I mostly worked on the frontend with Aaryan using Tailwind CSS for creation and customization of multiple pages, and provided backend support for them. The team was split into 2 frontend and 2 backend developers.

Kevin Mammadli
I primarily focused on the backend, where I implemented an algorithm for parsing and querying essential data from Purdue's dining court API. Additionally, I contributed to the development of the Preferences page, gathering necessary inputs about users.

Youngjun Yoo
I worked on the back-end. It was my first time using an algorithm that includes Purdue’s Dining Court API and OpenAI’s Chat Completion API. In addition, I created the landing page containing the sign-up/sign-in button.

Yashvardhan Khaitan
Aaryan Gautam

Updates

Youngjun Yoo started this project — Jan 21, 2024 07:10 AM EST

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.