With recent questions about inflation of the USD and significant institutional adoption of Bitcoin, we see cryptocurrencies as an important and exciting part of our future. As everyday people look to invest their money in these products, many are turning to the influencers on Twitter for buy and sell advice. We have seen tweets from the likes of Elon Musk cause meaningful swings in the market price of cryptocurrencies and we wanted to make access to this sort of influential information more easily accessible to everyday investors.
What it does
Our product allows our users to input the Twitter usernames of influencers that they wish to follow, pulls recent Tweets and then opens a stream which watches for any new tweets from that influencer. These tweets are stored in a database which the product uses to display a stream of tweets and a color-coded indicator for whether the tweet's sentiment is bullish or bearish which the user can use as a buy or sell indicator. Alongside the tweet stream, the user will see a historic price chart for the cryptocurrency as well as plots of historic sentiment.
How we built it
For this project we decided to use Python, PyQt, sqlite3, matlab plots, Twitter's API, and Coingecko's API. We planned to build this program using an MVC file format to keep our UI, logic, and backend data separate and easy to work with as we progressed each day with the project.
Challenges we ran into
Due to the nature of the project, we were faced with constant challenges of trying to put all the aspects of the project together, especially when it came to connecting the UI to the backend. However, as a group we tackled each error/bug one at a time to slowly build a functional project. Last but not least, we also we were testing massive amounts of data from the Twitter API and Coingecko API, which added a level of complexity when it came time for testing and debugging. All in all, the challenges we faced only accelerated our learning and sharpened our skills for the future.
Accomplishments that we're proud of
Throughout the entire project we were proud of the fact we built a relatively difficult application in a very limited time, while taking into account application scalability, massive tech upscaling, utilizing other technical tools that we haven't used prior to this hackathon. Although, we didn't execute everything perfectly - it has set us up to be more flexible when it comes to developing other applications in the future.
What we learned
As a team we have learned so many different tips and tricks from each other that we probably wouldn't have learned from sitting in class. For example, we implemented an MVC design to organize our files into bite size components so it was easier to update and make changes to the program as we progressed. We also learned how to read API/language documentations to help gather and implement our ideas into working products. Notably, we learned how to communicate as developers of varying skill to ensure that all aspects of the project were being worked on and echoing any potential blockers a team member may have on their designated work stream to make sure everyone was aligned.
What's next for Memes to Dremes
We plan to continue to make this product a more usable and functional where the user will have an easier time getting the most of out of the project. We would also like to take this project as a web application so that all sorts of devices can access its capabilities. Additionally we would want to introduce a more robust backend system that could handle more information than our current backend system.