Senior Performance Engineer, Discrete GPU

About Sarvam

Sarvam is building the bedrock of Sovereign AI for India. The company is developing India’s full-stack sovereign AI platform, building across research, models, infrastructure and applications with a singular focus on making AI genuinely work for India. Sarvam works with leading enterprises and public institutions and is backed by Lightspeed, Peak XV, and Khosla Ventures. Sarvam partners with India’s leading brands, including Tata Capital, SBI Life, CRED, IDFC, and LIC.

About the Role

Own sarvam's discrete-GPU surface: NVIDIA RTX / RTX Pro via TensorRT and CUDA, AMD ROCm-supported GPUs via ONNX Runtime, and DirectML as a Windows-universal fallback. The workstation form factor is the most relaxed footprint target but the most demanding latency target - users running creative tools alongside sarvam’s products expect zero perceptible cost.

What You’ll Do

• Land Sarvam’s edge models on RTX / RTX Pro / AMD GPUs, hitting predefined SLAs with headroom for concurrent GPU workloads.

• Author TensorRT plugins and custom CUDA / ROCm kernels where stock ops don't hit budget.

• Drive workstation device CI.

What We're Looking For

• 5+ years on ML deployment with 2+ years on discrete GPU inference.

• Production TensorRT including plugin authoring and engine refit.

• CUDA fluency at minimum at debug level; kernel authoring is a strong plus.

• Nsight Systems / Compute fluency.

• ONNX Runtime + DirectML EP experience.

Bonus Points

• AMD ROCm / HIP production experience.

• CUDA kernel authoring.

Why Sarvam?

Sarvam is a fast-moving, high talent-density team building full-stack AI for India, working on problems that push the frontiers of AI with real population-scale impact.

• Work alongside researchers, engineers, builders, and business leaders who move fast and hold each other to a very high bar

• High ownership and high impact, from day one

• Everything we do is AI-first, from the way we build and ship to the way we think about problems

• You can work on problems that could change how an entire country learns, works, and communicates

If you want to work on problems at the frontier of AI in India, Sarvam is the place to be.