• Designed and deployed scalable, cloud-native workspace environments for AI workloads using Kubernetes APIs, Kustomize, and Python on AWS EKS.

• Developed GPU auto-scaling solutions for EKS clusters, optimizing costs and resource utilization.

• Automated end-to-end CI/CD processes using GitLab CI, Jenkins, and Docker.

• Streamlined cloud infrastructure management with Terraform, ensuring reliability and scalability.

• Built Helm charts and Ansible playbooks to automate Kubernetes-based service deployments.

• Developed MLOps solutions using Kubeflow to automate machine learning pipelines.

• Enhanced scalability of ML workloads with custom Kubernetes operators and Go-based automation scripts.

• Created Docker-based simulation software for satellite communications, integrating backend systems.

• Conducted MLPerf benchmarks, optimizing TensorFlow GPU workloads with NVIDIA toolkit integrations.

• Developed a compiler for ONNX models using the Glow framework, extending support for TensorFlow models.

• Integrated PMD (Program Mistake Detector) into the development workflow, automating resource leak detection and enhancing system reliability during development.

• Designed and implemented a Single Sign-On (SSO) framework using OAuth2, enhancing security and user experience across platforms.

• Automated system documentation processes, improving efficiency and reducing errors in generating Interface Control Documents (ICDs).