Cura

Inspiration

Our inspiration for Cura came from personal experiences within our team. Some of us grew up with curly hair in families that had no knowledge of how to properly care for it. This led to confusion and frustration around which products to use and how to manage our hair effectively. We realized that many people, particularly women, face the same challenges with their unique hair textures. Cura was created to empower individuals with personalized hair care guidance, eliminating the guesswork and helping them embrace their natural beauty.

What it does

Cura is an AI-powered platform that provides personalized hair care and styling recommendations based on your specific hair texture. By analyzing individual hair types, Cura suggests the best products and routines to help users manage their hair confidently. Whether you have curly, straight, or textured hair, Cura tailors its recommendations to your unique needs, ensuring that every day can be a good hair day.

How we built it

We built Cura using HTML, CSS, and JavaScript for the frontend to create an intuitive and responsive user interface. For the backend and AI model training, we used Python. We experimented with Google Cloud Vertex AI and OpenAI to develop and fine-tune our large language model (LLM). We also utilized Nyckel for additional AI training. Our dataset, sourced from this GitHub repository, provided comprehensive data on various hair types and textures, which we used to train our model.

Challenges we ran into

One of our main challenges was experimenting with multiple AI platforms like Google Cloud Vertex AI and OpenAI, figuring out which one best suited our needs for accurate hair texture analysis. Another challenge was ensuring the model was trained on diverse and relevant data to provide meaningful recommendations. Additionally, integrating these AI technologies with our frontend while maintaining a seamless user experience was a technical hurdle we had to overcome.

Accomplishments that we're proud of

We’re proud of successfully building a functional platform that combines beauty and AI to address a real need. Experimenting with and integrating advanced technologies like Google Cloud Vertex AI, OpenAI, and Nyckel into a cohesive system was a major accomplishment. We’re also proud of the accessibility and simplicity of the platform, ensuring that users can easily navigate and benefit from personalized hair care recommendations.

What we learned

Through this project, we deepened our knowledge of AI platforms, especially Google Cloud Vertex AI and OpenAI, learning how to fine-tune models for specific tasks like hair texture analysis. We also gained insight into the hair care challenges people face and how technology can help solve these issues. On the development side, we honed our skills in creating user-friendly interfaces that work smoothly with AI-driven backend systems.

What's next for Cura

Next, we plan to enhance Cura’s AI by further refining our models using more diverse datasets and increasing the accuracy of hair care recommendations. We also aim to add new features like live consultations, a product recommendation engine, and partnerships with beauty brands for personalized product suggestions. Our ultimate goal is to expand Cura into a full-service platform for hair care, offering comprehensive styling advice and long-term hair health plans.

Built With

• css
• flask
• github
• google-cloud-vertex-ai
• html
• javascript
• llm
• machine-learning
• nyckel
• openai
• python