Staff Data Scientist, Imaging

About this role

<div class="content-intro"><p>Biohub is the first large-scale initiative bringing frontier AI models, massive compute, and frontier experimental capabilities under one roof. We're building a general-purpose system to accelerate scientific discovery, integrating frontier AI models, biological foundation models, and lab capabilities, with the ultimate goal of curing disease. Our technology powers scientists around the world, translating AI capabilities into tools that accelerate research everywhere.</p></div><h2><span style="color: rgb(40, 40, 39); font-family: helvetica, arial, sans-serif;">The Team</span></h2> <p>Our AI research team sits at the heart of our mission to unlock new dimensions of biological understanding. You will leverage state-of-the-art AI to accelerate discovery and drive transformative insights in biology—developing novel AI models purpose-built for biological research, engineering robust systems that enable breakthrough science at unprecedented scale, and translating these advances into practical tools that empower researchers worldwide.</p> <p>Our approach is comprehensive and integrated, bringing together world-class AI model development, exceptional engineering talent, high-quality biological data, powerful computing infrastructure, and strategic partnerships. Success requires excellence across five interconnected pillars: training frontier AI models specifically for biology; building engineering systems that maximize research velocity and efficiency; executing a sophisticated data strategy that fuels AI development; operating a world-class AI compute platform; and creating impactful products that transform AI capabilities into accessible scientific tools.</p> <h2><span style="color: rgb(40, 40, 39); font-family: helvetica, arial, sans-serif;">The Opportunity</span></h2> <p>This role is part of the Data team, which focuses on owning the strategy, sourcing and implementation for data supporting AI research and development. Our goal is to maximize the speed, agility, and capability of biological AI research by connecting public data resources and Biohub's experimental platforms to AI systems.</p> <p>The data that trains biological frontier models comes in dozens of modalities—sequences, images, spatial coordinates, time series, molecular structures, metadata, preprints and published papers—each with its own noise characteristics, biases, and information content. The question of how to represent this data for learning is one of the most important open problems in biological AI.</p> <p>You will operate with broad scope and high autonomy, influencing roadmap decisions across teams while mentoring senior individual contributors. Success in this role means scaling data systems that are not only large, but adaptive, interpretable, and scientifically grounded, accelerating progress toward robust biological frontier models and ultimately advancing human health.</p> <p>We're looking for data scientists who can work at this frontier: people who understand biological measurement deeply, think creatively about data representations and tokenization strategies, and can translate that thinking into novel training architectures. You'll work directly with experimental and computational scientists, data scientists and&nbsp; AI researchers to define what the models see and how they see it, and data engineers to make this work at scale. This is a role for someone who wants to invent the methods that make biological frontier models possible.</p> <h2><span style="color: rgb(40, 40, 39); font-family: helvetica, arial, sans-serif;">What You'll Do</span></h2> <ul> <li>Design data representations and tokenization strategies for imaging data that enable novel model architectures</li> <li style="font-family: helvetica, arial, sans-serif;"> <p>Coordinate Experimental, Data Science, Data Engineering and AI Research teams to translate biological structure into learnable representations—defining priorities and appropriate structures for metadata and data that information models can access and consume</p> </li> <li style="font-family: helvetica, arial, sans-serif;"> <p>Work across those teams to guide data acquisition priorities, define quality criteria, and assess external datasets from a representation perspective</p> </li> <li style="font-family: helvetica, arial, sans-serif;"> <p>Develop and validate approaches for combining heterogeneous data modalities into unified training frameworks, designing for robustness to noise, bias, and batch effects</p> </li> <li style="font-family: helvetica, arial, sans-serif;"> <p>Evaluate how representation choices impact model performance, identifying which biological signals are captured or lost and iterating to improve</p> </li> </ul> <h2><span style="color: rgb(40, 40, 39); font-family: helvetica, arial, sans-serif;">What You'll Bring</span></h2> <ul> <li style="font-family: helvetica, arial, sans-serif;">PhD in computational biology, bioinformatics, or a quantitative biological field</li> <li style="font-family: helvetica, arial, sans-serif;">Experience with tokenization strategies for non-text data (images, sequences, graphs, time series)</li> <li style="font-family: helvetica, arial, sans-serif;">Track record of novel methodological contributions (publications, open-source tools, or production systems)</li> <li style="font-family: helvetica, arial, sans-serif;">Familiarity with biological foundation models (ESM, scGPT, or similar)</li> <li style="font-family: helvetica, arial, sans-serif;">Deep understanding of imaging data, their underlying data characteristics, and how to transform raw data into ai-ready datasets.</li> <li style="font-family: helvetica, arial, sans-serif;">Experience designing data representations or feature engineering for machine learning, ideally in scientific or biological contexts</li> <li style="font-family: helvetica, arial, sans-serif;">Familiarity with modern ML architectures (transformers, diffusion models, or similar) and how