Efficiently find people who are your fans/supporters on Twitter. All you have to do is tag @myprofilevibes and it will show the top 3 supporters.

I worked on all aspects of this project. Starting from the idea, brainstorming, designing, developing, maintaining, and marketing. Checkout Twitter Page

Hurdles I faced

1. Who is your fan is a very subjective question. It differs for each person. Writing a generalized algorithm was hard.

How I resolved them

1. Launched a v1 version of the algorithm and gathered responses from the users. Based on it wrote a new graph algorithm that efficiently (reducing API usage) finds the perfect fan list. This algorithm took into account everything from the normalized activity of each follower to the frequency of activities. The final algorithm was verified by many users.

 Great features

1. The backend is developed on FastAPI. User asks for fan -> Twitter stream listener -> Goes to backend -> Generates an image using Python -> Retweets with that image. The image generation based on the template was a great feature.

Stack

Frontend: Next.js

Backend: FastAPI + Twitter APIs + Image processing

Hosting: Heroku (dynos might be low)