# AI Platform Engineer, Training and Inference at Saviynt

- Company: Saviynt
- What the company does: Discover how Saviynt secures all identities, enabling enterprises to manage access, governance, and compliance confidently while addressing the unique security challenges of the AI era.
- Company website: https://saviynt.com
- Type: Startups (AI role)
- Level: Mid level
- Location: Milpitas, California
- Work setup: Hybrid
- Pay: $274K to $304K base salary per year (USD)
- Posted: 2026-05-18
- Apply by: 2026-10-08
- Apply: https://jobs.lever.co/saviynt/9a8661ce-8856-4977-87f4-b06567125e28
- Page: https://www.1752.vc/careers/jobs/saviynt-ai-platform-engineer-training-and-inference/

## About the role

• Build and operate the LLM inference mesh with Ray Serve: compose vLLM (PagedAttention), SGLang (RadixAttention), and NVIDIA Triton (TensorRT/ONNX) as a unified deployment graph with Plasma zero-copy memory sharing

## What they're looking for

- • Experience in ML engineering with time in an ML platform or MLOps role
- • Production Ray depth: Ray Train, Serve, Core, and Data — debugged real production failures including NCCL timeouts, Plasma OOM, and Serve autoscaling lag
- • LLM serving engines: hands-on with vLLM, SGLang, or NVIDIA Triton — PagedAttention, prefix caching, and continuous batching tuned for latency/throughput targets
- • Distributed training: DDP, FSDP, NCCL collectives, gradient checkpointing, and mixed precision (BF16/FP8)
- • RL working knowledge: PPO, policy gradient, or RLHF — able to translate an algorithm into distributed compute primitives
- • Model lifecycle operations: MLflow registry, shadow/A/B/canary patterns, and auto-

Tags: Platform Upgrade