Join Sarvam in building AI solutions at scale for India's diverse population. We are seeking an Infrastructure Engineer to manage and enhance the core systems that power our AI products. This role is crucial for maintaining our Kubernetes clusters across multiple cloud environments and ensuring the seamless operation of our network, identity, and delivery pipelines.

At Sarvam, engineers own their systems end-to-end, with a strong emphasis on automating manual tasks. We integrate AI agents into our operations to assist with planning, incident triaging, and routine issue resolution, always with engineers providing the necessary oversight.

This position offers the opportunity to work on cutting-edge AI products, including voice agents, coding assistants, and our API platform, all while contributing to the foundational infrastructure that makes them possible.

As an Infrastructure Engineer, you will be instrumental in keeping Sarvam's AI product infrastructure running smoothly. This includes managing daily operations such as deployments, provisioning, access control, and addressing cost-related inquiries.

You will be expected to automate recurring manual tasks, transforming them into efficient pipelines, Terraform modules, templates, or self-service solutions for product teams.

A key aspect of this role involves building AI-assisted automations, such as agents that can triage alerts, summarize incidents, review routine changes, or answer common infrastructure questions, with human approval for critical actions.

Core responsibilities also encompass security best practices, including patching images and dependencies, credential rotation, and resolving security findings. You will maintain system reliability through accurate dashboards and alerts, track SLOs, manage costs, and eliminate noisy alerts. Participation in on-call rotations is also a requirement.

We are looking for candidates with 2–5 years of experience in DevOps, SRE, infrastructure, or backend engineering. A proven track record of building infrastructure independently—whether through a homelab, open-source contributions, or side projects—is highly valued, and you should be prepared to showcase your work.

Proficiency in Linux command-line operations, including process management, networking basics, logging, and systemd, is essential. You should have working knowledge of Docker and Kubernetes, with experience deploying and debugging applications on a cluster.

Strong scripting skills in Python, Go, or Bash are required, along with basic experience in Git and CI/CD practices. You must be able to demonstrate instances where you've identified and automated repetitive manual tasks, explaining the impact and savings.

Experience using LLMs or AI agents for automating real-world tasks (beyond just code writing) is a significant plus. We value engineers who fix root causes rather than recurring symptoms and proactively question manual processes.

Ideal candidates are curious, take ownership, and are eager to grow into platform engineering roles. Familiarity with Terraform, public clouds (AWS, Azure, GCP), Prometheus/Grafana, and exposure to SIP, WebRTC, or real-time media are considered advantageous, as are any cloud or Kubernetes certifications.