Inspiration
We often find ourselves shopping for an item online and end up finding multiple products of the same item which seem interesting. This leads to an overload of items in our shopping cart and we end up having a hard time deciding which is the better option. While we do have the options to read through the endless reviews and very single features, it can be exhausting and boring. This is why we wanted to create an extension which allows us to get instant feedback to maximize our cart while minimizing our costs.
What it does
Scans your cart and asks your preferences for each item to suggest which is a better option.
How we built it
We built this browser extension using HTML, CSS, and JavaScript. The popup interface was created with HTML and styled using CSS. JavaScript handles dynamic functionality like adding, managing, and deleting categories and preferences. The manifest.json configures the extension, and a content script interacts with specific web pages. Finally, we tested and refined the extension in the browser's developer mode.
Challenges we ran into
One of the biggest challenges was how we were going to collect reviews. During our research phase we discovered several options of scraping, such as adeify , untangled and firecrawl. They would scrape the whole site yet it would collect a lot of garbage data. addinial y the start up cost for these services were large enough to deter us from taking a gamble on a product that could be more an issue then an asset.
Additionally several of our target sites have protections against bots, such as session tokens, cookies and captcha would stop the crawler mid execution throwing several errors.
To overcome this issue we decided to use an in-place scraper so we can collect and clean our data, to then feed into our reasoning algorithm.
Accomplishments that we're proud of
What we learned
Although this wasn't our group's first experience with working in a fast paced group environment, we still took more than what we could have asked from this event. For one most of our team's background was in web and app development yet for the sake of the challenge we decided to take on the task of building a chrome extension. Due to the team aspect we sharped our git skills. Additionally we experimented with our prompt engineering and learned how to use docker for our, and narrowly were able to deploy on adeify
What's next for Shopmaxx
We want to add a feature where instead of analyzing the cart, we want it to get instant feedback on a product itself when you are on the page of the product.
Built With
- css
- html
- javascript
- json
- openai
Log in or sign up for Devpost to join the conversation.