Sanket Shelke

AWS Cloud Operations engineer with nearly 2 years running production infrastructure across 8+ client accounts, now moving into DevOps and Site Reliability Engineering. AWS Certified Solutions Architect – Associate, with day-to-day work spanning CloudWatch monitoring, EventBridge event-driven alerting, Python/Bash automation, patching, IAM, and cost optimization.

Portfolio backs the direction: a GenAI AIOps agent on Bedrock, CI/CD to ECS Fargate, and a self-healing EKS platform, all built with Terraform and GitHub Actions.

Associate Consultant – Cloud Operations at Yash Technologies (2024-12 – Present)

• Monitored and triaged CloudWatch alarms across an 8+ account AWS environment, resolving incidents against defined SLAs and accelerating recovery through faster root-cause isolation; configured metric and log alarms with standardized thresholds and notification routing so issues surface before user impact.
• Engineered proactive alerting pipelines using Amazon EventBridge and SNS to capture AWS Health and scheduled-maintenance events, delivering automated email notifications that surfaced service-impacting activity ahead of downtime.
• Automated recurring housekeeping and health-check routines with Python (Boto3) and Bash, eliminating repetitive manual effort and freeing shift capacity for remediation work.
• Built Microsoft Copilot automation agents to generate weekly status reports across client accounts, cutting a recurring 2+ hour daily reporting task to under 15 minutes and reducing manual operational overhead.
• Managed AMI lifecycle end to end (on-demand images, daily automated backups, retention, cleanup), removing stale AMIs and orphaned snapshots to reduce EBS storage cost without weakening recovery readiness.
• Ran monthly OS patching across ~15–20 Linux and Windows servers using SSM Patch Manager with maintenance windows, holding patch compliance with zero unplanned downtime, and performed in-place Windows Server OS version upgrades with pre-upgrade AMI backups and post-upgrade validation.
• Provisioned AWS Site-to-Site VPN connectivity (customer gateways, virtual private gateways, route propagation, and vendor configuration handoff to on-prem teams) and encrypted EBS volumes with AWS KMS for secure, compliant client environments.
• Administered IAM users, roles, and least-privilege policies, and managed security groups, route tables, and SSH connectivity on client request for controlled, secure network access; supported a cross-region workload migration (me-south-1 to eu-west-3) with AMI replication and post-migration validation.
• Built an advisory AIOps agent that detects incidents from CloudWatch alarms and runs RAG-based root-cause analysis over runbooks with an AWS Bedrock model, returning a ranked, evidence-backed remediation plan to on-call engineers.
• Kept it read-only with a human approval step (recommends, never executes) and provisioned the event-driven stack (Step Functions, Lambda, Bedrock Knowledge Bases, Guardrails) via Terraform and GitHub Actions, cutting simulated triage from ~20 minutes to under a minute.
• Containerized a Spring Boot REST API and shipped it through a GitHub Actions pipeline that tests, builds, scans, and pushes images to ECR, then deploys to ECS Fargate behind an Application Load Balancer.
• Provisioned VPC, interface/gateway endpoints, and services with Terraform; debugged VPC endpoint, S3 gateway, and ALB health-check timing issues to reach reliable zero-downtime deployments.
• Deployed a Spring Boot service to Amazon EKS with multi-replica Deployments monitored via Prometheus and Grafana, plus a Horizontal Pod Autoscaler that self-healed failed pods and scaled on CPU load.

B.E. in Computer Engineering – SKNCOE, Pune (2020 – 2024)