Lighthouse AI

Inspiration

We were inspired by the need to make healthcare accessible to low-income individuals who often face challenges navigating insurance plans. By leveraging AI, we aimed to simplify this process and provide real-time support.

What it does

Lighthouse AI suggests personalized insurance plans based on user inputs through a form-focused AI tool. Additionally, it features a RAG-based chatbot to answer any insurance-related questions, ensuring patients get real-time support. It also shares patient information with doctors, reducing redundancy in care.

How we built it

We used LangChain and OpenAI to develop the form-based AI recommendation system and integrated a RAG-based chatbot for real-time interaction. Pinecone was used for vector storage, and we implemented a secure, efficient flow for sharing patient data with doctors.

Challenges we ran into

We faced challenges in fine-tuning the AI to provide accurate, personalized recommendations and ensuring seamless integration of the chatbot with the backend. Managing patient data securely while preventing redundancy was also critical.

Accomplishments that we're proud of

We are proud of developing an AI-driven system that not only simplifies insurance selection but also provides real-time support to underserved populations, improving accessibility to healthcare.

What we learned

We learned how to effectively combine AI-driven solutions with real-world healthcare needs, ensuring that our tools are user-friendly, scalable, and secure.

What's next for Lighthouse AI

We plan to further enhance the AI's capabilities by integrating more insurance options, refining the chatbot's accuracy, and expanding the system to support additional healthcare services.

Built With

• amazon-web-services
• langchain
• nextjs
• node.js
• pinecone
• python
• react
• typescript