Tushar Verma

Results-driven DevSecOps/SRE Engineer with 3 years of experience designing, automating, and optimizing cloud-native infrastructure on AWS. Proven expertise in Kubernetes ecosystem tools including EKS, Karpenter, Kyverno, Argo Workflows, and Argo CD, with hands-on experience in Terraform, Crossplane, and Helm for infrastructure provisioning and GitOps-driven cluster add-on management. Skilled in driving cost optimization, EKS cluster lifecycle management including upgrades across 40+ clusters, and disaster recovery planning and execution.

Adept at building monitoring and alerting solutions using Dynatrace, Splunk, and OpenTelemetry, leveraging Python for automation and Ansible for configuration management to reduce manual intervention and improve system reliability.

DevSecOps Engineer at ADP (2024-06 – Present)

• Led a cloud cost optimization initiative by architecting automated EKS cluster scale-down during weekends and non-business hours across lower environments using Argo Workflows, achieving ~$7,000 monthly reduction in cloud spend while maintaining zero impact on developer productivity.
• Implemented Karpenter on the production cluster, configuring NodePools and EC2NodeClasses for workload-driven provisioning; enabled bin-packing consolidation and automatic deprovisioning of underutilized nodes, and set up dedicated node pools per namespace — significantly reducing total node count from 192 to ~140, lowering compute costs through right-sized, dynamic scaling.
• Migrated custom admission webhook to Kyverno ClusterPolicies for automated injection of tolerations and node affinity rules, ensuring deterministic pod scheduling onto Karpenter-managed node pools with zero workload disruption.
• Managed Amazon EKS cluster upgrades across 40+ clusters (version 1.35) spanning lower environments to production, provisioned via Crossplane and Terraform, ensuring minimal downtime and workload compatibility; maintained cluster add-ons through Helm charts integrated with Argo CD, aligned with security and compliance standards.
• Executed annual Disaster Recovery (DR) exercise in us-west-2, provisioning RDS, OpenSearch, and ElastiCache clusters using Terraform; restored OpenSearch via snapshot with ~10Cr searchable documents, orchestrated full service bring-up of a parallel site, achieving 4-hour RTO in compliance with organizational DR standards.
• Migrated from Splunk Connect for Kubernetes (SCK) to Splunk OpenTelemetry Connector (OTel) using ArgoCD and Helm, optimizing data flow, automating deployment, and ensuring seamless telemetry integration with Splunk.
• Conducted a POC evaluating EKS Pod Identity as a replacement for IRSA (IAM Roles for Service Accounts), assessing simplified IAM credential management, reduced operational overhead, and improved scalability for pod-level AWS authentication.

Site Reliability Engineer at ADP (2023-09 – 2024-06)

• Spearheaded infrastructure automation and deployment activities on production, ensuring seamless operations and resolving critical issues effectively.
• Orchestrated end-to-end RHEL migration, server provisioning, and environment setup to streamline application deployments across critical environments; wrote Ansible playbooks to automate migration tasks and configuration management, reducing manual effort by 60%.
• Automated consolidation of alerts from Splunk and Dynatrace using Python, Splunk REST API, Jenkins, and Crontab, sending consolidated incident reports per product via Webex bot; implemented FastAPI to streamline bulk issue closure, significantly reducing manual intervention for SREs.
• Contributed to the Windows Server migration project, successfully transitioning systems from Windows Server 2012 to 2016 ahead of end-of-life deadlines, ensuring continuity, compliance, and improved performance.
• Configured Dynatrace monitoring with management zones, alert profiles, and dashboards for a newly onboarded product; built dedicated dashboards for Enterprise Redis and AWS MSK to track critical metrics during upgrades and patching, contributing to 99.99% uptime.
• Conducted a POC using OpenTelemetry (OTEL) Collector to replace OneAgent for telemetry data collection, enabling vendor-neutral observability and enhancing flexibility and scalability.
• Implemented Ansible-driven automated weekend service restarts across production and UAT environments, improving system reliability, reducing manual intervention, and ensuring consistent uptime across critical services.

Bachelor of Technology in Information Technology – GL Bajaj Institute of Technology and Management (2019-08 – 2023-08)