ML-DL

Predicted demand using sales figures, mobility data and several demographic and footfall variables

RL-based challenge to robustly maintain an electrical grid without disruptions against an adverse agent. Achieved $28^{th}$ rank in the hackathon

Extracted HTML UI elements from wireframe drawings of websites ideating novel multi-pass inference technique to boost recall. Achieved $3^{rd}$ rank in the hackathon.

Explored Active Learning with different querying strategies on CIFAR10 dataset and managed to achieve high accuracies with very limited training data

Implemented and optimised multiple live models (eg. action recognition on something something dataset with over 200 classes) on edge devices (eg. Jetson TX2, Raspberry Pi)

Performed damaged car parts segmentation for automating auto claims with additional features like explainability and automated claim / damage report generation