Hands-on experience building evaluation and testing systems for AI/ML or LLM-powered products: eval harnesses, golden datasets, LLM-as-judge, regression testing across model and prompt changes, red-teaming, and guardrail validation. You will define what "high quality" means for an AI-native healthcare company — building the testing frameworks, agentic quality systems, and evaluation infrastructure that let us release faster with more confidence, and that earn clinical and enterprise trust in AI-driven features.