Principal Data Scientist, Science

The Chan Zuckerberg Initiative was founded by Priscilla Chan and Mark Zuckerberg in 2015 to help solve some of society’s toughest challenges — from eradicating disease and improving education to addressing the needs of our local communities. Our mission is to build a more inclusive, just, and healthy future for everyone.

The Team

CZI supports the science and technology that will make it possible to help scientists cure, prevent, or manage all diseases by the end of this century. While this may seem like an audacious goal, in the last 100 years, biomedical science has made tremendous strides in understanding biological systems, advancing human health, and treating disease. 

Achieving our mission will only be possible if scientists are able to better understand human biology. To that end, we have identified four grand challenges that will unlock the mysteries of the cell and how cells interact within systems — paving the way for new discoveries that will change medicine in the decades that follow:

• Building an AI-based virtual cell model to predict and understand cellular behavior
• Developing state-of-the-art imaging systems to observe living cells in action
• Instrumenting tissues to better understand inflammation, a key driver of many diseases
• Engineering and harnessing the immune system for early detection, prevention, and treatment of disease

CZI’s work in science includes grantmaking programs, open-source software development, and close collaboration with the Chan Zuckerberg Biohub Network. The CZ Biohub Network includes the San Francisco, Chicago, and New York Biohubs as well as the Chan Zuckerberg Imaging Institute. CZI also collaborates with institutional partners like the Kempner Institute for the Study of Natural & Artificial Intelligence at Harvard University. Join us in accelerating science.

The Opportunity

The Principal Data Scientist will lead a team that will define and create the datasets and covariates required to train a Virtual Cell model that understands how each cell type functions at a molecular level under healthy conditions, and is able to predict the impact on cells and cell populations of genetic or environmental perturbations. You and your team will partner closely with our Science Program team to partner or generate the required datasets, with Data Engineering and ML Engineering to decide data formats, schemas, and access patterns, and with AI Research to define the annotations, covariates, and quality measures required to use the data. 

CZI manages and processes scientific datasets specifically designed to enable biological modeling. We handle over 100 million fully standardized unique cells worth of single cell transcriptomic data, over 15 thousand cryoET tomograms that are in imaging datasets as large as 20TB and counting. This year, we are expanding data operations by > 10x to support a higher volume of imaging, sequencing, literature, and mass spectrometry datasets. These data are available via public resources, CELLxGENE Discover and CryoET Portal. Our resources provide access to open source data that is structured and used by tens of thousands of scientists each month to quickly query and form hypotheses on understanding how genetic variants in cells impact disease risk, define drug toxicities, and eventually discover better therapies.

As the Principal Data Scientist, you will manage a team of data scientists who are embedded in cross-functional, AI-modeling focused teams and are responsible for leading dataset definition and delivery to support modeling work. Success is measured by the speed with which datasets are available, the ease with which they can be used, the suitability of the dataset to address the modeling task, and the clarity with which data quality is available to the research team. As a manager of a small team, this role will be a player/coach position. Your individual contributions will focus on defining a strategic, integrated dataset design sufficient to realize a Virtual Cell Model, and to ensure that the individual datasets your team creates for each modeling project each create measurable progress towards the integrated design. 

What You'll Do

• Manage a team of data scientists to create high quality, standardized datasets for biological AI modeling and open-source publishing.
• Establish systems and processes to ensure agility and responsiveness to emerging needs from the rapidly evolving AI biology space.
• Define a data vision for the Virtual Cell Model, build buy-in with cross-functional partners, and partner to see it incrementally realized as we iterate towards our vision.
• Discover and define new data generation opportunities, and manage the delivery of those data products to our AI team.
• Collabor