• Led initiative to scale organizations core infrastructure and software development lifescycle using AWS Lambda, EventBridge, Sagemaker and Step Functions to ensure scalability and limited fault tolerance

• Proposed and developed a low-latency Python based data pipeline that served content to global users using S3, EC2, and PostgreSQL

• Developed an impedance controller using ROS, C++, Python, PyTorch, Scikit-Learn, Pandas, and Fast Robotic Interface (FRI) that enhances overall performance of coupled human-robot systems by utilizing Autoregressive Neural Network for future coupled state predictions. Deployed and tested it on a 7-DOF KUKA LBR iiwa820 robotic arm.

• Implemented state machines using Python library SMACH to execute complex robot behavior, drive the experiment protocol and explicitly control state transitions across the entire ROS framework.

• Developed alternate prediction methods traditionally used in pHRI like circle fitting and linear fitting to emphasize the robustness and accuracy of the proposed controller. The controller and framework achieved a 25% increase in stability and 40% increase in overall performance, quantified by overshoot, mean speed, smoothness, user effort and restoring force.

• Assisted Dr Hyunglae Lee (Spring '23) and Dr Wanxin Jin (Fall '23) in conduting Robotics and Autonomous System core course MAE 547 Modelling and Control of Robots

• Spearheaded global upgrades of commercial payment orchestration system’s microservices by implementing API rate limiters, message queues to ensure reliability, scalability and high throughput using Python, Java, SQL, JavaScript, GCP, BIS.

• Led initiative as Subject Matter Expert of payment microservices and applications to optimize AppDynamics-hosted applications, reducing thread consumption by 60% and log file sizes by 50% across all applications and services.