Inspiration
Growing up in Jamaica as a young Black girl, I struggled to understand how to properly care for my natural hair. Like many others, I wasn't taught about moisture retention, protective styling, or building a healthy hair care routine. By high school, years of heat styling and tight hairstyles had left my hair significantly damaged, and I had no idea how to identify the cause or fix it.
As I became more involved in online communities for Black hair care, I realized my experience was far from unique. Every day, thousands of people ask questions like:
Why is my hair breaking? Why are my edges thinning? Which products actually work? How often should I wash or deep condition my hair?
Most advice comes from trial and error or social media, making it difficult to know what applies to an individual's hair.
I wanted to build a tool that makes expert guidance more accessible and personalized for people with textured hair.
What it does
coil is an AI-powered hair care companion designed specifically for people with textured hair.
Users create a personalized hair profile by providing information about their hair type, concerns, goals, lifestyle, current products, and hair history. They can then upload photos of their hair for AI-powered analysis to identify potential issues such as:
Heat damage Split ends Dryness Breakage Color or chemical damage
Using this information, coil generates personalized recommendations and routines tailored to each user's unique needs.
Unlike existing solutions, coil is designed to become a long-term hair care companion rather than simply diagnosing a problem.
How we built it
How We Built It
The project was built using:
GPT-5.6 for intelligent reasoning and structured recommendation generation OpenAI Codex throughout development for planning, architecture, debugging, project scaffolding, code generation, and troubleshooting Next.js for the frontend FastAPI for backend services Supabase for authentication, database management, and Row Level Security Tailwind CSS for the user interface
Codex dramatically accelerated development. Instead of spending hours creating project scaffolding, configuring authentication, debugging dependency issues, or organizing the architecture, I could focus on designing a product that solves a real problem. It also helped generate implementation milestones, review code, troubleshoot deployment issues, and maintain development momentum throughout the hackathon.
Challenges we ran into
Building a full-stack AI application within a hackathon timeframe presented several challenges.
Some of the biggest hurdles included:
configuring authentication with Supabase applying database migrations correctly setting up environment variables deploying a multi-service application recovering development progress after an interrupted Codex session balancing ambitious product ideas with what could realistically be completed during the event
Another challenge was prioritizing features. There were many ideas that would improve the experience, but selecting the highest-impact functionality was essential for delivering a working prototype.
What we learned
This project reinforced that successful AI products require much more than simply connecting to a language model.
Building an effective experience also involves thoughtful user experience design, secure backend architecture, structured data models, authentication, and creating workflows that users will continue to engage with over time.
I also gained valuable experience using GPT-5.6 and Codex as development tools. This was my first time using Codex. Rather than replacing software engineering, they significantly accelerated planning, implementation, debugging, and documentation, allowing me to spend more time refining the product itself.
Future development
coil has significant room to grow beyond this initial prototype.
Future features include:
AI-powered progress tracking using before-and-after photos Hair growth analytics and progress visualization Calendar reminders for wash days, deep conditioning, moisturizing, and protective styles Personalized hairstyle recommendations based on hair goals and current condition A preferred hairstylist network so users can discover providers that specialize in textured hair Product recommendations informed by community discussions and reviews Adaptive recommendations that continuously improve based on user feedback Expanded support for additional hair textures, scalp conditions, and ingredient compatibility
Ultimately, the vision for coil is to become a personalized AI hair care coach that grows alongside each user's hair journey.
Why this matters
Textured hair has historically been underserved by technology, despite millions of people seeking personalized guidance every day.
My goal with coil is to make expert hair care knowledge more accessible, reduce the trial-and-error many people experience, and empower users to build healthier relationships with their hair through personalized, AI-driven recommendations.
This project began with my own experience growing up in Jamaica, but I hope it can eventually help people with textured hair around the world.
Log in or sign up for Devpost to join the conversation.