Latest AI JobsBrowse Jobs
Sarvam AI logo

Infrastructure SRE - HPC

Sarvam AIBengaluruInfrastructure
Apply on company site
Salary
Not listed
Job type
Full-time
Posted
26 Jun 2026
Status
Open · checked 7d ago

About this role

Sarvam AI is seeking an Infrastructure SRE - HPC in Bengaluru for a full-time position focused on managing large-scale, multi-vendor GPU environments. This opportunity is ideal for seasoned reliability engineers with deep domain expertise in GPU clusters, high-performance storage, or advanced networking who want to build foundational sovereign AI technology for India.

Role Overview

Sarvam AI runs a massive, multi-vendor GPU infrastructure supporting uninterrupted multi-week training jobs alongside latency-sensitive inference services. This is an advanced infrastructure and reliability position requiring deep technical ownership beyond standard administration. Professionals will manage complex challenges spanning parallel filesystems, RDMA fabrics, NCCL hangs, and distributed training failures. The organization employs specialists across specific focus areas while maintaining collaborative support across the entire fleet.

Key Responsibilities

Manage the GPU fleet from end to end, covering training and serving environments across provisioning, observability, capacity planning, and overall fleet health.
Participate in active on-call rotations, author resilient runbooks, and conduct thorough postmortems to implement permanent system improvements.
Design and develop custom internal tooling rather than relying solely on standard off-the-shelf management systems.
Collaborate closely with machine learning and platform engineering teams to maintain the stability of large training runs and ensure predictable serving latency.

Requirements

Minimum of five years of professional experience in infrastructure or site reliability engineering.
At least two years of direct experience operating GPU clusters at scale, though exceptional specialists in storage or fabric domains with specialized backgrounds are also encouraged to apply.
Proven ownership of critical infrastructure on-call duties, accompanied by a history of postmortems that drove meaningful systemic changes.
Strong programming proficiency in Python or Go for building and maintaining internal operational tooling.
Working fluency across multiple infrastructure domains to effectively identify, triage, and route incidents.

Nice to Have

Experience managing hybrid environments combining Slurm and Kubernetes.
Background in on-premise GPU deployments, including collaboration with data center operations on power, cooling, and InfiniBand cabling.
Familiarity with Indian NCPs, DGX SuperPOD, Lambda, CoreWeave, or NeevCloud.
Practical experience with multi-tenant GPU isolation methods such as MIG, MPS, or time-slicing in production environments.

Skills and Tools

Python
Kubernetes
Distributed High-Performance Storage
InfiniBand and RoCE Fabric Networking
GPU Systems Reliability and NCCL Debugging

About the Company

Sarvam AI is developing India's comprehensive full-stack sovereign AI platform, encompassing research, foundational models, specialized infrastructure, and practical applications. The company partners with major brands and public institutions to make artificial intelligence genuinely effective for India. Supported by prominent investors including Lightspeed, Peak XV, and Khosla Ventures, Sarvam AI offers a high-talent environment focused on population-scale impact.

How to Apply

Interested candidates should visit the official Sarvam AI careers page to submit an application and indicate their primary area of technical focus.

Skills and tech stack

PythonKubernetes

Questions candidates ask

Where is this job located?

This full-time position is located in Bengaluru, India.

What experience level is required for this role?

Candidates must have at least 5 years of experience in infrastructure or site reliability engineering, including a minimum of 2 years operating GPU clusters at scale.

What programming languages are required?

Proficiency in Python or Go is required for building and maintaining internal tooling.

Is remote work available for this position?

No, this is an on-site role based in Bengaluru.

How should I apply for this position?

Candidates should apply directly on the official company careers page and indicate the specific area of focus that matches their experience.

Ready to apply?

The application is on the official Sarvam AI careers page.

Apply on company site