Join Sarvam's Work Agents team and contribute to building the essential infrastructure for developing, evaluating, and deploying autonomous agents at scale. This role offers a unique opportunity to work across both infrastructure engineering and operations. You will develop self-service tooling and automated pipelines, while also ensuring the smooth day-to-day operation of CI/CD, environments, networks, and deployments.

This position requires a hands-on DevOps engineer with expertise in Kubernetes, cloud infrastructure (primarily AWS, secondarily Azure), and a strong understanding of networking and CI/CD for multi-cluster, multi-tenant environments. Collaborate closely with harness development engineers to make their work shippable, observable, and operable.

Key responsibilities include managing multi-cluster Kubernetes environments (upgrades, node lifecycle, RBAC, policy enforcement), owning CI/CD pipelines from commit to production (build, test, scan, sign, deploy with rollout patterns), provisioning and maintaining cloud resources via Infrastructure-as-Code (IaC) with cost awareness, managing platform-level network plumbing (CNI, ingress, routing, policies, VPNs), building automation and tooling for self-service operations, maintaining the observability pipeline (metrics, logging, tracing), and ensuring reproducible environments through IaC tools like Terraform.

You will also be involved in multi-vendor cluster bring-up and maintaining the overall health and performance of the platform.

We are seeking candidates with 3+ years of experience in DevOps, SRE, or platform/infrastructure engineering, demonstrating hands-on ownership of systems. Fluency in Kubernetes operations and debugging is essential. Proficiency in AWS (VPC, EC2, EKS, IAM, storage, networking) with working knowledge of Azure is required. A deep practical understanding of networking concepts (CNI, ingress, service routing, network policies, VPNs/tunneling) is crucial.

Experience building and maintaining CI/CD pipelines (GitLab CI, GitHub Actions, Argo CD, Flux) and strong DevOps tooling knowledge are necessary. We prefer candidates with strong software engineering fundamentals in Python or Go, and a product mindset focused on internal user adoption and reduced toil.