Senior/Principal Machine Learning (ML) Engineer- Oracle Health AI- REMOTE

Building off our Cloud momentum, Oracle has formed a new organization - Oracle Health. This team will focus on product development and product strategy for Oracle Health, while building out a complete platform supporting modernized, automated healthcare. This is a net new line of business, constructed with an entrepreneurial spirit that promotes an energetic and creative environment.

In this highly strategic role, you will drive the design, development, and integration of state-of-the-art AI solutions into healthcare products as an applied science leader. You will collaborate with stakeholders across product, engineering to shape the Healthcare AI roadmap and deliver value to customers and partners.

As ML Engineer, you will drive the scientific vision for Healthcare AI systems.

Responsibilities

• Build and optimize LLM-powered agents for clinical and operational workflows (summarization, prior auth, patient messaging, coding).
• Select and version foundation models and prompts to meet privacy, cost, latency, and safety targets.
• Add guardrails, uncertainty handling, human-in-the-loop, and evidence-grounded citations.
• Adapt LLMs with healthcare data via prompt engineering, instruction tuning, and preference optimization.
• Evaluate methods using health-related datasets to ensure outputs are accurate, reliable, and trustworthy.
• Implement RAG, structured outputs, and function/tool calling for EHR and partner integrations.
• Maintain evaluation harnesses and datasets; run error analysis and active learning to reduce bias and close gaps.
• Productionize services with CI/CD, containerization, autoscaling, vector stores, and cost controls.
• Instrument traces, metrics, and logs across prompts, tools, retrieval, and outputs; enforce SLAs with canary/blue-green rollouts and safe rollback.

Qualifications

• Have an MS/PhD in Computer Science, ML, or related field, or equivalent experience
• 2+ years production experience with LLM agents
• Proven record improving agent reliability via reward modeling, policy learning, constitutional methods, tool-use strategies, and safety
• Publications, open-source contributions, or patents in LLMs, IR, or clinical NLP.
• Productionize services with CI/CD, containerization, autoscaling, vector stores, and cost controls.
• Instrument traces, metrics, and logs across prompts, tools, retrieval, and outputs; enforce SLAs with canary/blue-green rollouts and safe rollback.
• Productionize services with CI/CD, containerization, autoscaling, vector stores, and cost controls.
• Instrument traces, metrics, and logs across prompts, tools, retrieval, and outputs; enforce SLAs with canary/blue-green rollouts and safe rollback.