HairBase | Devpost

Welcome Screen in the Onboarding View
One of the Preferences Pages in the Onboarding View
Home + Search Screen
Product View Page — with Preference Matching
Camera Screen — Uses Groq and OCR to find a Product from a Photo
Camera Screen — Matches for Given Product, Option to add a new Product

Inspiration

We wanted a tool to help shoppers understand and differentiate the hundreds of hair products available.

What it does

HairBase is a mobile application designed to help users identify what hair care products would be most effective for them to use. With a wide rage of hair type, texture, color, and consumer conscientious, HairBase makes it easy for people to find the hair produces made for them and their needs.

How we built it

We built HairBase with a React Native frontend, a Python FastAPI backend, and a PostgreSQL database. The backend utilizes Groq for OCR and product information retrieval. The API and database were deployed on a self-hosted server using Coolify that allowed our app to access our own endpoints to receive data.

Challenges we ran into

One challenge we ran into was achieving consistency and accurate output from the AI model with varied input. We solved this challenge by refining our prompts and constricting inputs and outputs where possible.

Accomplishments that we're proud of

We're proud of aggregating data for thousands of products and integrating Groq's api with a clean UI and UX. It was everyone's first time using AI vision models for text recognition, and we're proud to be able to implement these new technologies in such a short time frame. Additionally, we reverse engineered Sephora's website in order to seed our own product database with nearly 2000 items.

What we learned

Each member in our group went into this Hackathon knowing a different aspect about what would make our app work, and we all learned a little bit more about how the systems that we're used to building work with each other and share share information. It was super interesting to see how all the components in a full-stack app work together and being able to create a small scale version of that opened our eyes a lot.

What's next for HairBase

HairBase could be expanded to include more retailers and more types of products to eventually become a truly personalized shopping experience. Our computer vision and AI integration also have potential applications to assist those with vision disabilities or deal with language barriers.