We were curious about big name brands that we see every day in the grocery stores. We may know what Pepsi is, but we don't know what other brands they own or where they are located. We hope to one day build out a platform where consumers are more conscious about what companies they support in hopes that they would buy more locally and invest locally.
What it does
Allows users to look up info about a particular brand and learn more about that brand. Maybe there is a brand that is marketed as "sustainable" but ends up being a big company instead that doesn't really follow sustainable practices.
How we built it
We built it by divide and conquer method. Linden did the front-end, Chi worked on the backend.
We wanted to spin something up fast and used an SPA. So we went with React.js for the front end and Express as the web framework. We used MongoDB to host the data and Mongoose as the ORM.
Challenges we ran into
We originally wanted to look for a third party API that had supply chain data and other important information like carbon footprints for big brands, but couldn't find anything. We found one by Alphawave, but it wasn't free.
Another challenge was integrating the Rust machine learning into Node.js with WebAssembly, and processing the image files uploaded from the front end.
Accomplishments that we're proud of
We both work on weekdays and weekends, so another challenge was working in real time with each other, but we honestly worked very well given the challenge of being online at different times. We are proud of being able to make an idea happen in a very short period of time. An accomplishment is being able to query the data on the client side. It definitely is always nice to see that data displayed on the page.
We got the initial version of image upload/ML recognition working, but we will need better ML models to extract relevant info from photos.
What we learned
We learned a lot about Blue Diamond Growers and the pros and cons of consuming Almonds! :P Working efficiently isn't easy when the idea isn't clear. At the same time, you have to be willing to go through the vagueness of an idea. Asking questions is important to allow yourself to fully understand what your teammates mean when they talk about a feature/user story. That way, there aren't as many hiccups while building.
What's next for KnowTheBrand
There is a lot more to do.
- UI/UX is a big one. We want this to be a smooth experience for users
- Re-architecting the data model. We might want to add more information than just brand names. We would love to curate news about specific brands when a user looks something up.
- For fun! Machine Learning. This would be better as a mobile app, since most people are on their phones and use it when they are shopping. We'd hope to have users take a photo of a product as a way to query information about the brand.