SniffSniff

SniffSniff & Sniffodoo will help you sniff out the right link to your class’s discord/slack and allow you to skip reading through hundreds of unread messages by summarizing important information.

Comment

Inspiration

Our inspiration is from just our everyday lives and seeing other students' pain points. As classes are transitioning back to in-person, students are having a harder time finding the discord/slack link without people pasting it into the zoom chat during lectures. Often it is not until the 5th week a student accidentally stumbles upon the link and realized they’ve been missing out on the community of classmates they’ve been relying on.

What it does

SniffSniff helps sniff out the right discord/link with just a few clicks. If the channels have not been created for certain classes the user is trying to find, the user can create one and the link of the channel created will automatically upload it to our database so other classmates can sniff you out too. Sniffodoo is a bot that helps sniff out important information from those slack/discord channels and allows you to skip all the hundreds of unread messages by summarizing important information such as homework tips, exam dates/prep, and the most critical information.

How we built it

We first designed a wireframe for SniffSniff’s website through Figma, and then we built our website using Next.js and CSS tools. In order to garner data about classes and instructors for the new semester, we web-scraped berkeleytime.com using Selenium and Requests packages. Subsequently, we imported our data into the backend of our website. The bot was built in python using Discord.py to scrape data from a channel and then using co:here’s NLP API to analyze, summarize and classify the data. We used a lot of different kind of models and endpoints from co:here's API to accomplish the functionality including text summarizers, information extractors, and sentiment analyzers. We had to construct the data for the prompts using real-life examples from the discord classes we're in.

Challenges we ran into

We had a hard time figuring out new technology tools that were offered, such as co:here and cockroach. Furthermore, we had some doubts about what project we were going for in the beginning, but we quickly overcame that. Other challenges we also ran into include figuring out how to scrape the berkeleytime website, along with some design difficulties with CSS. The biggest challenge, however, was trying to stay sane and awake.

Accomplishments that we're proud of

We are proud of everything we were able to accomplish in such a short amount of time. We have never done a project of such a scale in 24 hours, and we are also proud of how we were able to bond with one another and learn from each other during the process.

What we learned

We learned that with time, good teamwork, and a whole lot of effort, amazing creations can be made within 24 hours. It really shows as long as you start and keep pushing, you can create anything!

What's next for SniffSniff

Next for SniffSniff we definitely want to optimize Sniffodoo (our bot) to process information faster when it is called. We also want to create an email service using Sniffodoo to send out a summary of important information daily to ensure all students stay caught up without putting in too much effort.

Built With

Share this project:

Updates