Staff Machine Learning Scientist, Agentic AI

POSITION SUMMARY:

Natera is seeking a Staff Machine Learning Scientist – Agentic AI to join our AI team, an advanced R&D and core AI innovation team bridging the gap between molecular discovery and clinical execution. Leveraging a proprietary data moat of over 250,000 oncology patients profiled with longitudinal ctDNA, WES/WGS, digital pathology, and EMR data, you will design and deploy production-grade autonomous AI agents and multi-modal foundation models. Your mission is to architect systems capable of multi-step biological reasoning, converting complex multi-omic datasets into verifiable clinical insights that accelerate biomarker and therapeutic discovery. You will lead the next evolution of our Agentic AI platform, designing autonomous systems capable of reasoning through the complexities of cancer biology, orchestrating proprietary foundation models, and simulating virtual patient trajectories.

PRIMARY RESPONSIBILITIES:

• Lead the technical design and deployment of multi-agent systems capable of autonomous hypothesis generation and tool use, including genomic variant calling, LLM fine-tuning, and clinical trial matching pipelines
• Incorporate and advance Natera’s transformer-based foundation model by integrating DNA, RNA, and H&E imaging modalities for multi-step biological reasoning and tool use
• Implement advanced LLM reasoning frameworks, such as ReAct and Chain-of-Thought, alongside reinforcement fine-tuning (RFT) to ensure agents provide accurate, explainable clinical rationales
• Architect systems that autonomously translate complex, multi-modal data into diagnostic and therapeutic insights with human-verifiable reasoning and tracing
• Own the technical strategy and product roadmap for agentic workflows across the Biopharma Solutions and Therapeutics Discovery division, converting complex clinical challenges into scalable AI systems
• Establish production-grade machine learning engineering standards and reproducible architectures across the AI team to ensure absolute model transparency and scientific auditability
• Drive cross-functional alignment and technical consensus by defending agentic architectures and biological reasoning frameworks in rigorous peer reviews

QUALIFICATIONS:

• PhD or Master's degree in Computer Science, Bioinformatics, Statistics, or a related quantitative field
• 8 or more years of experience in AI research or engineering, with a proven track record of moving multi-agent orchestration architectures or large-scale language model workflows from prototype to production
• Deep experience with agentic frameworks, such as LangChain or Claude Agent SDK, retrieval-augmented generation (RAG), and validation frameworks for autonomous AI agents
• Strong understanding of cancer genomics (WES/WTS), mutational signatures, and structure-activity relationships
• Advanced production-level development experience using PyTorch and experience with distributed training on large GPU clusters, including NVIDIA H100s

KNOWLEDGE, SKILLS, AND ABILITIES:

• Ability to operate with absolute ownership to close operational gaps and independently drive architectural deployment
• Data-driven decision-making focused on empirical model performance and clinical validity
• Technical leadership capability to define long-term AI engineering roadmaps
• Rigor in code architecture, reproducibility, and production-grade software engineering practices
• Comfort with high intellectual friction and the ability to defend scientific and engineering choices under rigorous internal peer review
• Focus on translating machine learning outcomes directly into patient-centric clinical utility