Latest AI JobsBrowse Jobs
Sarvam AI logo

Platform Engineer - AI Infrastructure

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

About this role

Sarvam AI in Bengaluru is seeking a Platform Engineer to focus on AI Infrastructure. This role involves building control-plane services, schedulers, and developer tooling for a massive GPU fleet, making it an ideal opportunity for experienced systems engineers passionate about sovereign AI.

Role Overview

Sarvam AI operates a large multi-vendor GPU fleet that simultaneously supports intensive training workloads and demanding inference services. This position focuses on the software building side of infrastructure rather than day-to-day operations. The selected professional will design and ship the scheduling, multi-tenancy, serving, and self-service layers that empower machine learning teams to utilize thousands of GPUs independently. Working in Bengaluru, the engineer will treat internal machine learning teams as customers and build maintainable platforms that minimize hardware contention and failure.

Key Responsibilities

Design and implement the serving platform control plane to convert model artifacts into scalable, multi-tenant endpoints with intelligent routing and rollout capabilities.
Build scaling and elasticity layers supporting both training and serving workloads, managing capacity pooling, queue-depth autoscaling, and bin-packing across graphics processors.
Develop scheduling and orchestration layers using tools like Kueue, Volcano, or custom controllers, emphasizing gang scheduling, priority preemption, and topology-aware placement.
Establish multi-tenancy, RBAC, and isolation models, exposing capabilities like MIG and time-slicing as self-service tiers with robust quota enforcement.
Manage networking components, including CNI configurations, ingress routing, RDMA exposure, and tenant network segmentation.
Create observability, cost tooling, and storage data-path abstractions to handle high-cardinality metrics and data-locality-aware placement.
Deliver developer experience tools such as CLIs, SDKs, APIs, and infrastructure-as-code control planes using Terraform, Crossplane, or custom operators.

Requirements

Five or more years of professional experience building infrastructure or platform software, demonstrating a strong history of designing shipped services and control planes.
Proficiency in software engineering using Go or Python, with the ability to build and debug complex production systems.
Deep expertise in Kubernetes at the controller and internals level, including writing operators or controllers and understanding API machinery.
Working knowledge of GPU-specific platform constraints, such as MIG, GPU sharing, gang scheduling, and topology-aware placement.
A product-oriented mindset focused on internal user adoption, clean API design, and enabling self-service workflows.

Nice to Have

Background in developing serving, inference, or training platforms at scale for model endpoints.
Hands-on experience with GPU scheduling tools like Kueue, Volcano, Slurm, or Run:ai in production.
Proven track record of shipping multi-tenant isolation features like MIG, MPS, or time-slicing as self-service capabilities.
Advanced familiarity with Kubernetes networking, CNI internals, RDMA, or SR-IOV configurations.
Prior experience working on on-premise GPU platforms or multi-vendor environments.
Contributions to open-source projects related to Kubernetes, scheduling, or GPU infrastructure.

Skills and Tools

Python
Kubernetes
Terraform
Go
Control Planes
GPU Scheduling Systems

About the Company

Sarvam AI is constructing India's full-stack sovereign AI platform, integrating research, models, infrastructure, and applications to make artificial intelligence truly effective for the region. The organization collaborates with major enterprises and public institutions, backed by prominent venture firms including Lightspeed, Peak XV, and Khosla Ventures, alongside brand partners such as Tata Capital, SBI Life, CRED, IDFC, and LIC.

How to Apply

Interested candidates should visit the official Sarvam AI careers page to submit their application for the Platform Engineer - AI Infrastructure role.

Skills and tech stack

PythonKubernetesTerraform

Questions candidates ask

What is the primary focus of the Platform Engineer role at Sarvam AI?

The role focuses on building the software control-plane, scheduling, multi-tenancy, serving, and self-service layers that sit on top of a large GPU fleet, enabling machine learning teams to utilize infrastructure without manual intervention.

Where is this position located?

This position is located in Bengaluru, India.

What level of experience is required for this position?

Candidates must have 5+ years of experience building infrastructure or platform software.

What programming languages and core technologies are expected?

Candidates should have strong software engineering skills in Go or Python, and deep expertise in Kubernetes at the controller and internals level, along with Terraform familiarity.

Does this role involve operations or site reliability engineering?

No, this is strictly the build side of infrastructure rather than the operate side. SREs handle fleet reliability and carry pagers, while this role builds the systems that make the fleet usable and resilient.

Ready to apply?

The application is on the official Sarvam AI careers page.

Apply on company site