Computational Pathology Scientist

仕事内容

Gilead's mission is to discover, develop, and deliver therapies that will improve the lives of patients with life-threatening illnesses

worldwide.

The Computational Scientist, Pathobiology will be responsible for the following:

Job Description

The Research Pathobiology group at Gilead Sciences is seeking a talented and highly motivated imaging data scientist to advance discovery research and development projects at our Foster City, CA headquarters.

As part of the computational pathology team within Research Pathobiology, you will support imaging biomarker development efforts for clinical drug applications by:

• Analyzing pathology imaging data (e.g., H&E, IHC, CISH, mIF, CODEX, Spatial Transcriptomics) generated across Gilead’s drug development pipeline.
• Developing image analysis tools using commercial, internal, and open-source packages.
• Building automated image analysis pipelines for deployment on-premises and in cloud-based high-performance computing (HPC) environments.

The role requires extensive cross-functional collaboration with pathologists, biologists, biomarker scientists, data and imaging scientists, and IT personnel.

The primary focus is supporting image analysis initiatives, with an emphasis on developing and applying novel deep learning and machine learning approaches to advance our understanding of pathobiology in oncology, virology, fibrosis, and inflammation.

The candidate will also partner with clinical imaging and data management teams to deploy, maintain, and integrate computational solutions on HPC in the cloud.

Essential Duties and Job Functions:

- Support the development of advanced analytics, computer vision, and computational tools to derive novel imaging-based biomarker endpoints.

- Collaborate with internal and external scientific partners to design, execute, and validate analytic strategies for tissue-based endpoints and imaging biomarkers supporting Gilead’s drug discovery and development pipelines.

- Evaluate and implement new computational approaches in digital pathology to extract histopathological endpoints and perform spatial analyses.

- Curate and prepare large imaging datasets for deep learning (DL) model training and development, including targeted models (tissue compartmentalization, cell phenotyping, etc.) and pathology foundation models.

- Contribute to imaging data management strategies and solutions within Gilead.

- Identify best practices and innovation opportunities relevant to image analysis projects.

- Effectively communicate findings and progress through reports, presentations, and publications for both expert and non-expert stakeholders.

Education/Basic Qualifications:

• PhD in a relevant quantitative field (e.g., Computer Science, Biomedical Engineering, Physics, Mathematics, Statistics); postdoctoral experience is a plus, OR
• MS degree in Computer Science/Biomedical Engineering with 4+ years of industry experience, OR
• BS degree in Computer Science/Biomedical Engineering with 6+ years of industry experience

Knowledge, Experience, and Skills:

• Proficiency in deep learning and data science libraries such as PyTorch, Pandas, scikit-learn and NumPy; experience with image processing packages such as OpenSlide, OpenCV, MONAI, or Elastix is a plus.
• Demonstrated expertise in Python for scientific computing and imaging data analysis; experience with additional programming languages is a plus.
• Extensive experience with DL models and architectures for image segmentation and classification such as ResNet, U-Net, and transformer-based models (e.g., ViT, Swin Transformer); familiarity with other ML algorithms (e.g., Logistic Regression, Random Forest, SVM)..
• Experience managing end-to-end ML/DL/AI projects, including data engineering, resource management, model training, selection, evaluation, and stakeholder communication.
• Up-to-date knowledge of advances in AI research and its application to medical imaging and digital pathology.
• Solid understanding of the mathematical and statistical foundations of machine learning and medical image analysis (e.g., optimization, image registration, segmentation, classification).
• Excellent written and verbal communication skills.
• Ability to multitask and prioritize while maintaining high standards of efficiency and quality.
• Self-motivated with a strong commitment to accuracy and excellence.

Preferred qualifications:

• Fluency in scientific computing environments (e.g., Unix/Linux shell), particularly in HPC and cloud-based clusters, is a plus.
• Publication record in deep learning, machine learning, or statistics, particularly in digital pathology, is a plus.
• Strong understanding of medical image data formats and challenges associated with large pathology images (e.g., WSI, CODEX, ST); experience analyzing whole-slide images is a plus.
• Experience with manipulating, analyzing, and visualizing large internal, public, and commercial imaging datasets is a plus.
• Familiarity with cell biology and microscopy is a plus.

People Leader Accountabilities:

•Create Inclusion - knowing the business value of diverse teams, modeling inclusion, and embedding the value of diversity in the

way they manage their teams.

•Develop Talent - understand the skills, experience, aspirations and potential of their employees and coach them on current

performance and future potential. They ensure employees are receiving the feedback and insight needed to grow, develop and

realize their purpose.

•Empower Teams - connect the team to the organization by aligning goals, purpose, and organizational objectives, and holding

them to account. They provide the support needed to remove barriers and connect their team to the broader ecosystem.