Staff/Principal AI Engineer, LLM Platform & Production

Zus Health

46 hours ago

Be an early applicant

Get a reply from this employer

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Health insurance

401k with match

Unlimited PTO

Wellness benefits

7+ years building and shipping AI/ML systems and 2+ years deploying LLM-powered products.
Hands‑on depth in modern LLM engineering: RAG, agents and tool use, structured outputs, prompt and context engineering.
Track record of building ML infrastructure: serving, pipelines, model and prompt versioning, CI/CD, monitoring.
Strong Python and software engineering fundamentals; fluency in PyTorch and the Hugging Face ecosystem.
Production experience on a major cloud (AWS) with containers and IaC.
Sound judgment about safety, privacy and reliability when models touch sensitive data.
Clear writing and communication; able to explain model behavior to clinicians, product leaders and executives.

Architect and own the LLM platform architecture, including model gateway, routing, prompts versioning, caching and guardrails.
Build agentic systems with tool-using agents that reason over patient data and surface grounded insights.
Own retrieval over clinical data: RAG, embedding strategies, vector/graph retrieval, grounding to sources.
Make evaluation a first-class system with offline/online evals, CI regression suites, and monitoring for drift and hallucination.
Lead model prompting strategies and fine-tuning (LoRA/PEFT, SFT) for clinical extraction and normalization.
Run AI/ML in production: CI/CD for models/prompts, inference serving, observability, human-in-the-loop feedback.
Prototype with providers and engineers to identify high-leverage problems and iterate on real usage.
Lead design reviews, document standards, and mentor engineers to keep Zus current with AI advances.