Research & Professional Experience

  • Completed an intensive 12-week AWS-funded cloud computing program covering AWS Cloud Foundations, Linux, Python, SQL, networking concepts including VPC design, subnetting, routing, and security protocols.
  • Deployed and managed EC2 instances, designed VPC architectures with public and private subnets, and implemented Infrastructure as Code using Terraform.
  • Worked hands-on with AWS services including EC2, S3, Lambda, IAM, CloudWatch, CloudTrail, EKS, and ECS, with automation using Python and GitHub Actions.
  • Built Python-based multi-task computer vision pipelines for segmentation, depth estimation, and surface normals using GitHub for version control and reproducible development.
  • Implemented experiment logging and monitoring with Weights & Biases, tracking metrics, hyperparameters, and model checkpoints for observability and data-driven iteration.
  • Orchestrated experiments on Linux GPU clusters using Slurm, optimizing resource utilization and applying semi-supervised learning to improve accuracy by +10.64% mIoU and +8.43% pixel accuracy.
  • Built Python-based 3D point cloud processing pipelines with experiment-driven workflows for adversarial attack analysis and defense evaluation.
  • Implemented logging and monitoring of experiments to track metrics, hyperparameters, and robustness results across Linux environments.
  • Optimized training and inference workflows on GPU clusters, improving execution efficiency by approximately 4%.
  • Designed and implemented an automated chemical dosing system using Arduino microcontroller programming for process control in industrial laundry machines.
  • Developed and integrated control algorithms to ensure consistent and precise chemical dispensing, enhancing reliability and repeatability of dosing cycles.
  • Improved overall process efficiency by reducing manual intervention and streamlining system operations.