Price Pal

• 

  Click to start analysis

• 

  Home page

Inspiration

We were inspired when browsing online marketplaces and having trouble

What it does

Instant market research for products from within the Stripe Dashboard

How we built it

We used FastAPI for the backend. We used Selenium and BeautifulSoup4 to take information about similar products from the web. We used the Stripe API and app framework to create our Stripe app.

Challenges we ran into

Retrieving product information proved to be too great a task for the limited time we had. We also wanted to fine-tune our own model to improve its abilities, but that proved to be too much too. We had major problems with over-scoping throughout the project.

Accomplishments that we're proud of

We wrote a multithreaded web scraper and pulled down tens of thousands of product examples for a chosen niche which was luxury watches. We were able to parse the HTML into JSON while scraping too, and we were left with robust categorical data for luxury watches. We were able to convert that into a single .CSV, but actually fine-tuning the model using that CSV proved to be too difficult, and that was the reason the project didn't come to fruition.

What we learned

We learned a lot about scope and that our idea, while great, was just a little bit too much for us to implement over the weekend. We also worked very well as a team, and learned a lot of development skills by combining our different backgrounds.

What's next for Price Pal

With another weekend, we could have an extremely useful tool for market analysis in small scale companies. With a year, we could integrate an LLM to ask the user for more information about their product to improve prediction and research accuracy.

Built With

• fastapi
• python
• react
• selenium
• stripe