Inspiration
Scandy AI was inspired by the need to make online shopping easier and more efficient. With so many product reviews and options available, users often struggle to quickly find the information that matters most. Our goal was to create a tool that summarizes reviews and tracks shopping carts automatically.
What it does
Scandy AI is a Chrome extension powered by Gemini 1.5 Flash and a summarization API. It generates concise summaries of product reviews from any shopping website, saving users time. Additionally, it can log items in your shopping cart across multiple sites into a Google Sheets for easy tracking.
How we built it
We built Scandy AI using advanced AI models and APIs to process product data and summarize reviews effectively. The Chrome extension integrates with these tools to extract relevant information from shopping websites. We also used Google Sheets API to log cart items for seamless tracking.
Challenges we ran into
One of the major challenges was prompt engineering to ensure accurate summaries of product reviews. Another challenge was extracting relevant page information from diverse shopping websites with varying structures. Both tasks required constant fine-tuning to provide a seamless experience.
Accomplishments that we're proud of
We are proud of successfully implementing a working summarization feature that saves users time. Additionally, we developed a reliable system to log cart items from different websites into Google Sheets automatically. These features make online shopping more efficient for users.
What we learned
We learned a lot about the complexities of web scraping and prompt engineering. Handling diverse webpage layouts and creating clear, concise summaries required us to adapt and experiment with different AI models and techniques. This project greatly improved our understanding of working with APIs and user data.
What's next for Scandy AI - Shop & Save
Next, we plan to enhance the product’s ability to summarize reviews more accurately and expand compatibility with more shopping websites. We also aim to integrate more personalized shopping features, such as tailored product recommendations and price tracking.
Built With
- gemini-ai
- google-sheets
- tailwind
Log in or sign up for Devpost to join the conversation.