About

I created a robot that autonomously accepts or rejects prospects on Tinder (a popular dating platform) based on physical appearance. This project was made under 24 hours during the Hack The U competition in 2017 (the largest hackathon event in utah). I was awarded second place and featured on the Mechanical Engineering Department’s website here.

I successfully used transfer learning on a convolutional neural network (CNN) to understand my preferences in female physical traits. CNNs are a category of neural networks that have proven very effective in identifying faces, objects and traffic signs apart from powering vision in robots and self-driving cars.

The Automated Tinder Bot works by first collecting an image of the phone screen through a webcam. It is then processed through the CNN that determines if the bot should swipe left (reject) or swipe right (accept). Once this is determined, MATLAB communicates with an arduino that uses a servo to swipe left or right. After swiping, the arduino communicates to MATLAB that it is done. The cycle repeats.

In the video, the capacitor placement on the servo was slightly off. I corrected this issue after the time of the video.