Senior Machine Learning Engineer
Core
Building SAGE, a real-time AI governance engine that monitors, governs, and remediates autonomous AI agents using custom small language models (SLMs) acting as judges in the live request path.
Role type
Senior Applied Machine Learning Engineer (Real-time AI Governance)
Builds
Production SLMs and classifiers for AI safety, Agent Rewind capabilities, and high-performance inference infrastructure for enterprise customers.
Domain
AI Safety, Autonomous Agent Governance, Real-time ML Systems
Deliverable
production ML models | product features | infrastructure
Required skills
End-to-end production ML ownership, Python, PyTorch, SFT, DPO/RLAIF/RLHF, vLLM/SGLang/TensorRT-LLM, continuous batching, KV-cache optimization, inference quantization, adversarial training, synthetic data generation, online/offline evaluation, drift detection, model failure diagnosis.
Preferred skills
Experience with LoRA, GRPO, speculative decoding, building automated red-teaming pipelines, mining long-tail violations from live customer environments.
Technologies
PyTorch, vLLM, SGLang, TensorRT-LLM, LoRA, DPO, RLAIF, RLHF, FP8/INT8 quantization, KV-cache, continuous batching, GRPO.
Responsibilities
Own full training lifecycle for SLMs including base-model selection, SFT, and preference optimization; design multi-stage inference pipelines handling real-time enforcement and batch workloads; build automated data curation and evaluation frameworks; diagnose model failures and optimize accuracy/latency/cost tradeoffs; partner with product and security teams to translate governance requirements into modeling problems.
Seniority
Senior, hands-on IC
