Senior Director, AI Diagnostics

Role overview

Lead the research and clinical translation of AI and computational methods for oncology diagnostics, guiding work from opportunity selection through validation, launch, and ongoing monitoring. The role sets a foundation model and multimodal research direction, turning longitudinal molecular, imaging, and clinical data into products with a clear clinical purpose.

Responsibilities

- Lead and expand an algorithm research team, recruit key talent, and build strong scientific and execution practices. - Define a research roadmap for foundation models and multimodal learning, selecting opportunities where these methods can create meaningful clinical value. - Establish milestones, reproducible analyses, and evidence-based decisions from feasibility through validation and launch. - Work with product, clinical, and commercial partners to prioritize use cases, define intended use and success measures, and decide whether work should advance, continue, or stop. - Guide rigorous data preparation, analytical and clinical validation, and post-launch performance monitoring. - Develop technical leaders and influence senior partners across research and development, quality, regulatory, clinical, and commercial functions.

Requirements

- Ph.D. in bioinformatics, computational biology, machine learning, biostatistics, or a related quantitative field; a master’s degree with substantial additional experience may qualify. - Senior-level experience leading multidisciplinary teams and complex computational product programs through clinical deployment. - Demonstrated ability to shape foundation model research strategy, including representation learning, pretraining, adaptation, and evaluation. - Accountability for taking at least one AI/ML or computational algorithm from research through validation and deployment in a diagnostic product. - Strong understanding of analytical and clinical validation for regulated diagnostics, including expectations for AI-based products. - Deep expertise in at least one oncology data type and sufficient breadth to review model design, data assumptions, evaluation, and Python-based analyses.

Nice to have

The role calls for leadership across molecular, imaging, and clinical data, with relevant examples including sequencing, liquid biopsy, transcriptomics, and whole-slide imaging.