Selected Projects

• Mock-Buddy, a web application that utilizes computer vision and deep learning to help users overcome stage fear and improve public speaking and presentation skills with real-time feedback. The application provides real-time feedback on facial expressions, body language, and voice modulation to help users become more confident and effective speakers.
• Quizzzy, a mobile application that uses Transformers to automatically generate multiple choice questions from context of textbooks. This provides fast and efficient way for students and teachers to self-evaluate and assess understanding, making it an ideal study tool.

• trAIner, a mobile application that uses pose estimation to monitor user’s activity during workout and provide real-time feedback on the correct form and technique. It helps users to perform exercises correctly and avoid injury, while also improving their overall workout experience.

• hard-hat-detection, a system that uses object detection model using deep learning to detect the presence of hard hats in images & videos, specifically for embedded systems. This model can be used to improve safety in construction sites by automatically identifying workers who are not wearing hard hats and alerting supervisors.

Selected Opensource Contributions

---

Repo	Keywords	PR#
ivy	Python - Tensorflow - MXNet - PyTorch - Jax - Numpy	7439, 7514, 7764
docus	Java - Android - Computer Vision	30
github-raven	Python - Slack - GitHub Integration	7
quich	C - CLI	15
HTML-Markdown-converter	Python - RegEx	8
FoodMenu	Unity - WebGL - Doc	20, 21