What you'll do
- Profile and tune vLLM/TensorRT/ONNX runtime for target SLAs
- Capacity planning, autoscaling, and cost–latency trade‑offs
- Implement caching and batching strategies; A/B changes in prod
What we're looking for
- Strong CUDA/Triton/TensorRT and GPU profiling
- Experience with serving frameworks and tracing
- Infra IaC skills; SLO/SLA mindset
Where this role fits
This role sits in the LLM Infra & MLOps TalentZone and commonly supports packs like Cost & Quality Sprint and Guardrail & Reg Audit. Work is routed under QA; this is a managed bench, not a marketplace.