Inspiration

Imagine you're planning for your retirement and need to browse products offered by Sun Life. Sun Life provides a wide range of products, but you might not have the time to perform exhaustive research. Instead, you may want a chatbot that recommends personalized products tailored to your specific needs.

What it does

Our chatbot provides a platform for users to interact with the Sun Life website and receive personalized recommendations for insurance and investment products. By understanding the user's unique needs, the chatbot streamlines the decision-making process and enhances user experience. Our chatbot runs a local GPT model that operates without a server, ensuring that all personal information is stored safely and remains private.

How we built it

We leveraged the following technologies:

  • React.js for the front end, ensuring a seamless and interactive user interface.
  • FastAPI for the backend, providing a lightweight and efficient framework for handling requests.
  • LangChain for building the conversational RAG (Retrieval-Augmented Generation) system, enabling intelligent and contextual responses.
  • Local GPT model that runs entirely on the user's device, enhancing security and privacy.

Challenges we ran into

  • Ensuring data privacy while handling user queries and recommendations.
  • Fine-tuning the AI model to provide accurate and personalized responses.
  • Integrating LangChain effectively with FastAPI for optimal performance.
  • Designing an intuitive UI that enhances user engagement and ease of access.

Accomplishments that we're proud of

  • Successfully implementing a privacy-first local AI chatbot that doesn't rely on external servers.
  • Building a robust recommendation system that tailors product suggestions to individual users.
  • Creating a smooth and user-friendly interface that integrates seamlessly with Sun Life's ecosystem.
  • Optimizing the chatbot for efficiency and fast response times using FastAPI and LangChain.
  • Deploying a local GPT model that operates entirely on-device, ensuring maximum security and privacy.

What we learned

  • The importance of on-device AI processing for privacy and security.
  • Best practices in combining AI with financial services to improve user experience.
  • Deep insights into conversational AI architectures and how to enhance chatbot intelligence.
  • Strategies for scalable and efficient web applications using modern frameworks like React.js and FastAPI.

What's next for SolGuard

  • Integration with Sun Life's internal chatbot to provide a unified and improved user experience.
  • Expanding features to automate claims processing and enhance convenience for users.
  • Enhancing AI capabilities to offer even more accurate and context-aware recommendations.
  • Incorporating voice support for a hands-free, accessible experience.

By continuously improving and expanding, our chatbot aims to make financial planning easier, faster, and more personalized than ever before.

Built With

Share this project:

Updates