Staff Research Engineer, SWORD Health

• Deep expertise in production ML systems - You’ve built and shipped ML systems that users depend on, with a track record of solving complex technical problems
• Production LLM experience - You’ve fine-tuned, evaluated, and deployed LLMs in production environments
• Experience with preference optimization - You’ve implemented techniques like RLHF, DPO, or similar approaches in production settings
• Strong ML fundamentals - You understand when to use different ML approaches, can design rigorous evaluation frameworks, and make principled technical tradeoffs
• End-to-end ownership - You’ve taken ambiguous, complex problems from initial exploration through production deployment and ongoing iteration
• Excellent engineering skills - You write production-quality code, understand distributed systems, and can debug complex ML pipelines
• Communication and influence - You can drive technical decisions across teams and explain complex tradeoffs to technical and clinical stakeholders

Desirable

• The Applied AI Team develops AI and ML solutions that power personalized clinical care at scale
• Our work includes end-to-end AI treatment management systems that handle complete care pathways - from initial assessment through intervention selection and progress monitoring across various therapeutic domains
• We are also building Phoenix, an AI Care Specialist that engages in ongoing dialogue with patients and provides clinical guidance throughout their recovery journey
• This work spans the full AI stack: prompt engineering, RAG implementations, model fine-tuning, and agentic workflows
• We leverage both generative AI and classical machine learning depending on what the problem requires
• All of our projects contribute to delivering digital care that reaches new levels of clinical quality and accessibility
• Own complex ML problems end-to-end - Take on the hardest technical challenges (like fine-tuning LLMs to maximize clinical outcomes) from initial exploration through production deployment and real-world iteration
• Drive technical strategy in your domain - Define how we approach model development, evaluation, and optimization for clinical AI applications
• Build and optimize production LLM systems - Design and implement fine-tuning pipelines, preference optimization (RLHF, DPO), and rigorous evaluation frameworks for clinical use cases
• Work deeply across the full AI stack - From data preparation and model training through deployment, monitoring, and iteration in production environments
• Influence across teams - Partner with Product, Clinical, and Engineering teams to shape technical decisions and ensure we’re solving the right problems
• Raise the technical bar - Mentor engineers, establish best practices, and help define what excellent ML engineering looks like at Sword
• May lead 1-2 engineers on specific projects, providing technical direction and mentorship

• 30-minute call with the recruiter
• 30-minute call with an initial screener
• 1 take-home assignment (to be delivered within 1 week)
• 1h general skills and working principles interview
• 1h mindset and cultural fit interview