Nischal Shrestha

AWS (IAM, EC2, ECS, VPC, S3, ALB, WAF, RDS, DocumentDB, ElastiCache), Terraform, Terragrunt, Ansible, Proxmox VE, Enterprise Networking Fundamentals

Docker, Kubernetes (K3s), Amazon ECS, EKS

GitHub Actions, Jenkins, CircleCI, Argo CD

Prometheus, Grafana, AWS CloudWatch, Amazon Athena (ALB Log Analysis)

Bash, Python, JavaScript

Trivy, Snyk, OWASP, ZAP

Hybrid GPU Compute Platform with Queue-Based Auto-Scaling

• Architected queue-driven auto-scaling GPU worker platform using RabbitMQ, AWS Lambda, CloudWatch, and ECS, enabling horizontal scaling based on workload demand.
• Built custom Lambda pipeline to monitor RabbitMQ queue depth and publish CloudWatch custom metrics, triggering ECS scaling via alarms.
• Deployed and optimized containerized CUDA GPU workloads on AWS ECS, troubleshooting CUDA/NVML runtime issues in Docker environments with Celery.
• Implemented custom EC2 AMI and user-data script bootstrapping to significantly reduce GPU worker startup time during scaling events.
• Designed hybrid ML compute architecture integrating on-premises GPU workstations with AWS cloud infrastructure, reducing compute cost while supporting long-running GPU workloads.

• Led planning, prioritization, and execution of DevOps initiatives, defining the team's long-term roadmap and growth strategy.
• Contributed to system architecture from early development stages, ensuring scalability, reliability, and alignment with engineering objectives.
• Directed migration of the entire system stack to a new platform with minimal downtime through structured risk mitigation strategies.
• Managed and mentored the DevOps team, fostering technical excellence, collaboration, and continuous professional development.
• Maintained sandbox environments using Kubernetes and Helm to enable rapid development and testing cycles.
• Designed contingency plans, release check protocols, and production readiness reviews to ensure system stability and deployment quality.
• Provisioned and tested on-premises infrastructure for hybrid-cloud environments using Proxmox VE and NAS, optimizing cloud infrastructure costs.

• Managed multi-environment AWS infrastructure using reusable Terraform modules and Terragrunt.
• Implemented secure CI/CD authentication using OIDC-based access for cloud resources.
• Developed reusable GitHub Actions workflows to standardize CI/CD pipelines across multiple projects.
• Analysed AWS ALB access logs using Athena and visualized metrics via Grafana dashboards.
• Improved cloud cost monitoring and optimized workflows for machine learning model deployments.
• Strengthened AWS security posture through IAM policies, roles, and granular access controls.
• Led stress testing and benchmarking initiatives to validate scaling policies and maintain system availability.

• Implemented Teleport to provide secure developer access to infrastructure with improved auditability. • Migrated CI pipelines from Jenkins to GitHub Actions, reducing deployment time by 50%. • Developed modular deployment pipelines supporting feature-based client configurations. • Provisioned AWS infrastructure (ECS, VPC, EC2, ALB, WAF, Security Groups, IAM, S3, Secrets Manager) using Terraform. • Mentored DevOps interns, enabling them to independently manage operational workloads. • Conducted vulnerability assessments and penetration testing (VAPT) across infrastructure, container images, and deployed systems.

• Developed Bash automation scripts and configured Jenkins pipelines. • Implemented monitoring and alerting using Prometheus, Grafana, and Alertmanager. • Created GitHub Actions workflows for automated issue tracking from commit messages. • Conducted load testing using k6. • Deployed and configured n8n for automation and backup workflows. • Automated Linux VM provisioning and configuration using Ansible. • Assisted in infrastructure migrations and early Kubernetes (K3s) deployments. • Deployed web applications on AWS using EC2, ALB, S3, and Security Groups. • Built foundational knowledge of Infrastructure as Code and CI/CD pipelines using Jenkins.