Prompt Safety ML Engineer - On-Device & Cloud Guardrails

Trilyon, Inc. in Mountain View, CA is seeking an experienced Machine Learning Engineer to lead the development of prompt injection and safety models for agentic AI systems across mobile, cloud, and XR/AR platforms.

You will design, train, and deploy classifier and guardrail models that screen inputs and outputs for injection attacks, unsafe content, and policy violations. A core part of the role is post-training these models with RLHF, DPO, and related optimization techniques to push detection