Fabulous Helper

Start View
Asking for Recommendation
Product Recommendation Result

Inspiration

Nowadays, finding the right electronic device can be challenging due to the overwhelming amount of options. Questions such as "Which laptop should I buy as an incoming student in [subject]?" or "I am planning to build my first gaming PC, what components shall I select to assemble with a budget of $3500?" frequently appear on forums like Reddit. Noticing these common dilemmas, we developed an AI-powered sales helper that offers germane recommendations based on user's needs and background information.

What it does

Similar to ChatGPT, our AI sales helper will provide users with recommendations for electronic devices, complete with images, links, a short description, and three key tags for each product, tailored to the user's specific needs and concerns.

How we built it

With a team of five people, we divided our group into subteams to tackle various tasks. These included building interactive components, crawling store pages, and integrating with OpenAI's API. Our toolset includes React, Express.js, Python, and the OpenAI API.

Challenges we ran into

Having officially registered for the competition in late June, we had approximately 10 days to brainstorm ideas and work on this project. Additionally, several team members were unavailable due to the July 4th holiday. Consequently, our team faced significant time constraints in coordinating and refining our work.

Accomplishments that we're proud of

We successfully implemented our AI-powered sales helper functionality in a remarkably short time frame. Despite the challenges, our team developed an interactive website that delivers highly accurate and personalized electronic device recommendations. This achievement not only demonstrates our technical prowess but also our commitment to enhancing user experience and solving real-world problems efficiently.

What we learned

We learned the importance of effective teamwork and collaboration, especially with teammates we had just met. By quickly understanding each other's strengths, we learned to leverage our individual skills to achieve our common goal. This experience taught us not only how to adapt and communicate effectively but also how to bring out the best in each team member, ensuring a successful and efficient project completion.

What's next for Fabulous Helper

We plan to extend our features to include recommendations for a wider range of products, beyond just electronic devices. Additionally, we aim to enhance our UI/UX to create a more beautiful and user-friendly interface. To continuously improve our recommendation system, we will experiment with various large language models (LLMs) and evaluate their performance to ensure we are providing the best possible experience for our users.