Staff Machine Learning Scientist, Translational AI

About the role

This role provides technical leadership at the intersection of deep learning, computational biology, and molecular diagnostics. You will guide the design and validation of sequence and multimodal models to advance patient stratification, target discovery, and treatment monitoring across our cell-free DNA and multi-omic platforms. This position offers significant technical autonomy to shape modeling strategies while also being hands-on in model development and testing.

Key facts

What you'll do

• Provide primary technical direction for deploying foundation models in molecular, genomic, and pathology applications for oncology and translational medicine.
• Develop rigorous workflows to align and fine-tune pre-trained foundation models using clinical trial and molecular diagnostic data, ensuring empirical grounding.
• Design and implement frameworks for evaluating model performance and providing constructive feedback to enhance modeling code, experimental standards, and scientific approaches within the AI research group.
• Lead the fine-tuning and evaluation of deep sequence, multimodal, and representation learning models for biomarker identification, recurrence monitoring, and treatment response prediction.
• Create workflows for fine-tuning, probing, and analyzing latent space representations to extract interpretable biological patterns from complex transformer models.
• Validate model outputs against multi-omic data and real-world outcomes to ensure deterministic accuracy for patient tracking and clinical decisions.
• Construct and optimize machine learning models using next-generation sequencing, ctDNA assay data, digital pathology images, and longitudinal clinical information.
• Design clinical investigation frameworks that link model performance metrics directly to translational utility and shifts in real-world data distributions.
• Identify and address algorithmic weaknesses, dataset biases, and covariate shifts, implementing effective solutions for regulated clinical pipelines.
• Collaborate with Computational Biology, Translational Science, and Medical Affairs teams to translate clinical needs into quantitative machine learning problems.
• Bridge AI Research and ML Engineering teams to ensure validation models transition smoothly into scalable production workflows.
• Offer technical guidance and data execution support for external collaborations and foundation model research initiatives.
• Prepare clear data packages detailing complex multimodal model architectures and performance for clinical governance and stakeholder updates.
• Lead the writing of technical papers for machine learning conferences and computational biology journals.
• Represent the company's translational AI capabilities at international medical, oncology, and machine learning events.

Requirements

• PhD in Computer Science, Computational Biology, Bioinformatics, Biomedical Engineering, or a closely related quantitative field.
• A minimum of 5 years of industry or post-doctoral experience applying deep learning to biological, genomic, or clinical data, with a specific focus on oncology or immunology.
• Strong technical understanding of transformer architectures, representation learning, self-supervised learning, or deep sequence modeling.
• Demonstrated ability to translate machine learning results into verifiable biological insights or clinical performance measures, beyond optimizing isolated metrics.
• Expert proficiency in PyTorch, HuggingFace ecosystem, parameter-efficient fine-tuning (PEFT) methods, Captum, MLflow, and distributed GPU computing.
• Proven technical leadership through end-to-end project ownership, architectural design, or guiding cross-functional teams.

Nice to have

• Experience building or fine-tuning multimodal foundation models that integrate genomic sequencing data with digital pathology images or electronic health records.
• Hands-on experience with clinical trial datasets, real-world data (RWD/RWE), or developing models for regulatory environments.
• A strong publication record as the primary author in leading machine learning venues.

Skills & tools

• Advanced mathematical and algorithmic understanding of deep learning, optimization, and probabilistic modeling.
• Ability to quickly learn complex cfDNA platforms, biochemistry workflows, and multi-omic data generation processes.
• Precise written and verbal communication skills with meticulous attention to algorithmic detail and statistical validation.
• Capability to manage independent projects while achieving cross-functional goals within matrixed teams.
• A proactive approach to balancing scientific rigor, operational speed, and resource constraints under demanding timelines.
• Experience utilizing cloud-based productivity and high-performance computing infrastructure.

Practical notes

• Compensation range: $163,200 - $220,000 USD.
• Benefits include comprehensive medical, dental, vision, life, and disability plans, free testing for employees and immediate families, fertility care benefits, pregnancy and baby bonding leave, 401k, and commuter benefits.
• Natera is an Equal Opportunity Employer committed to diversity and inclusion.
• All qualified applicants will be considered without regard to race, color, religion, gender, gender identity or expression, sexual orientation, national origin, genetics, age, veteran status, disability, or any other legally protected status.
• Natera will only contact candidates via email from an @natera.com domain. Be aware of potential recruitment scams.