Computational Biologist II, CellxState

Biohub•San Francisco, CA

7h•$153,000 - $210,100•Hybrid

About The Position

Biohub is launching the first large-scale initiative that integrates frontier AI models, massive compute, and frontier experimental capabilities to accelerate scientific discovery. We are building a general-purpose system to cure disease by integrating frontier AI models, biological foundation models, and lab capabilities. Our technology empowers scientists globally, translating AI capabilities into tools that accelerate research. The multi-dimensional imaging program builds imaging tools that capture life across scales, from single proteins to whole organisms, revealing how proteins and cells function, communicate, and assemble into living systems. These observations are forming the basis for a new generation of AI models capable of predicting cellular behavior and guiding the development of improved treatments for widespread diseases. This initiative unites Stanford, UC Berkeley, and UC San Francisco into a single collaborative technology and discovery engine. Our vision is to tackle significant scientific challenges that are not feasible in conventional settings, enable investigators to pursue their most innovative and risky ideas, and facilitate research for scientists and clinicians both within our home institutions and beyond. We are a team of passionate individuals driven by technology, guided by scientific research, and motivated by collaboration, all working towards the mission of curing or preventing all diseases. The CELLxSTATE Program, part of the Imaging Grand Challenge, is developing next-generation technologies to decode and control cellular decision-making. This involves combining live-cell imaging, multi-omics, and AI at an unprecedented scale. We also create extensive reference datasets, such as our OpenCell project (https://opencell.czbiohub.org/), which maps protein localization and interactions, for community-wide mining and reuse in discovery. Our scientific work is fully open-source and has been published in leading journals like Science, Nature Methods, and Cell (https://biohub.org/leonetti/publications/). At the heart of our current efforts is multiDPS (Multimodal Dynamic Pooled Screening), a high-throughput platform that integrates custom microscopy, automation, CRISPR screening, and molecular profiling to map and predict dynamic cell states. We are seeking a Computational Biologist to play a key role in image analysis for our next-generation Optical Pooled Screening program. This is an excellent opportunity for individuals with a strong interest in data science, engineering, and cell biology, who will be supported by experts in a highly collaborative and well-funded scientific environment. We champion team science, with our projects bringing together biologists, technology developers, engineers, data scientists, and AI/ML experts.

Requirements

PhD in Computational Biology, Biology, or Computer Science, or a MS with relevant job experience.
At least 4 years of experience in Python-based image analysis or scientific computing.
Fluency with computational tools and infrastructure such as Python, Github, and Slurm.
Experience with modern biological data formats such as OME-Zarr and AnnData.
Experience designing workflows for large, complex datasets, including scalable storage formats, and reliable metadata and experiment tracking.
A proven track record of individual innovation, together with a strong ability to work collaboratively.
A passion for research and understanding how cells work.
An enthusiasm for team science and open science - this position will be embedded in a large multi-disciplinary team.
Excellent written and oral communication skills.

Nice To Haves

Experience with fluorescence microscopy, confocal, or lightsheet is a plus.

Responsibilities

Design, develop, and maintain scalable image analysis pipelines for large-scale fluorescent microscopy datasets, with an emphasis on image quality robustness and computational efficiency.
Integrate emerging multi-modal data types (e.g., spatial transcriptomics) into unified, AI-ready datasets that support downstream modeling and discovery.
Advance our bio-image analysis capabilities (e.g., segmentation, tracking, image-stitching, and image registration).
Partner closely with biologists and automation engineers to implement end-to-end quality control metrics, ensuring the fidelity of our experimental and computational pipelines.
Architect modular and reusable processing frameworks that can flexibly support multiple experiment types within a shared infrastructure.
Publish and disseminate impactful findings through preprints, papers, and software repositories (e.g., GitHub).