This was our project for JunctionAsia 2022. We utilized the MetaVerse platform ZEP and integrated that with online education platforms like Coursera, EdX, and Udemy. Our problem definition was that online courses and degrees lack the interaction of real-time classes; hence, we tried to solve it usin

Designed and implemented ZEPademics, an online learning platform based on the idea of a virtual campus, in under 35 hours and presented it to the judges in the Hackathon. We were awarded as the most curious team among the 60+ participating teams in the Hackathon.

We had to replicate the paper CEM: Commonsense-aware Empathetic Response Generation(https://arxiv.org/abs/2109.05739) and also experiment with various ablation studies. In the end, we successfully replicated the paper and completed the ablation studies to get new insights and better understand the r

Working on a Deep Learning project to recognize human facial patterns independent of age, appearance, and ethnicity by using ResNet architecture with data augmentation along with training and validating on datasets- FER2013, CK+, JAFFE, and TFD. [Python, PyTorch].

Currently implementing a model to grasp and react as human with emotions in complex environments. Reproducing and proposing new methods to improve the EmpatheticDialogues paper (Facebook's research paper).