Staff Applied AI Engineer, Product & Agent Performance

📋 Description

• Own product-layer decisions shaping agent behavior, including prompting, retrieval, context
• Design retrieval and context architecture to deliver correct data to models across real healthcare
• Create memory/state handling for multi-turn and multi-agent flows, determining what is carried
• Develop context and prompt templates with few-shot examples and reasoning scaffolds for consistent
• Improve performance via prompting, tool-use strategy, and validated experimental approaches rather
• Build and run evaluations against real production conditions to measure performance, regressions

🎯 Requirements

• 8+ years of production software engineering, with 3+ years owning ML/LLM/agentic systems in
• Ability to diagnose agent failures and attribute fixes to instruction, retrieval, context, or
• Hands-on experience with RAG architecture, production-grounded evaluation frameworks, and
• Familiarity with AWS AI/ML services (Bedrock, SageMaker) for building and evaluating agentic
• Evidence-led judgment and the ability to push back on launch decisions; builder’s mindset to run

🎁 Benefits

• The opportunity to define how agent performance, safety, and readiness are measured for production
• Remote-friendly culture with ownership across prompts, context, memory, evaluations, and escalation
• Employee-driven programs and a diverse, purpose-driven Arcadian community.