Machine Learning Engineer (Model Dev)

We're looking for a machine learning engineer to help develop AI biomarkers that improve cancer care. You'll work closely with experienced ML scientists and engineers, as well as clinical, biostatistics, product, and regulatory partners, across the model-development lifecycle, from prototyping and experimentation through validation and production deployment. You'll contribute to challenging problems in medical AI, including building models from digital pathology and clinical data, improving robustness across scanners and sites, understanding model behavior, and advancing our pathology foundation models.

• Develop and evaluate AI-based biomarkers using multimodal data, including whole-slide images, clinical variables, and molecular data, to predict patient outcomes, treatment benefit, and molecular traits.

• Contribute to the development and evaluation of self-supervised foundation models and downstream machine-learning models, including multiple-instance learning, time-to-event / hazard models, segmentation, and classification.

• Develop and evaluate methods to improve model robustness and reproducibility across scanners, institutions, staining protocols, and patient populations.

• Explore and apply interpretability methods to explain model decisions, build clinician trust, and drive actionable model improvements.

• Build and improve tools and workflows that support efficient, reproducible model development, experimentation, validation, and deployment.

• Perform rigorous model evaluation and analysis, communicate findings clearly, and document experiments and technical decisions.

• Collaborate with ML scientists and engineers as well as product, biostatistics, clinical development, and regulatory/quality partners throughout model development and validation.

• Support regulatory and quality documentation related to AI model development and validation.

• Contribute to peer-reviewed publications, conference presentations, and external academic or industry collaborations.

• 1+ years of experience developing machine-learning or deep-learning models using PyTorch (or TensorFlow), including relevant master's or graduate research experience.

• Familiarity with oncology and biomarker development, including cancer biology and treatment pathways, clinical endpoints, risk stratification, and what makes a biomarker clinically actionable.

• Experience working with real-world datasets and evaluating machine-learning models using appropriate metrics and validation approaches.

• Strong Python programming skills and familiarity with modern software-development practices, including version control, testing, and code review.

• Ability to analyze experimental results, troubleshoot model behavior, and communicate findings clearly.

• Ability to collaborate effectively with ML engineers, scientists, and cross-functional partners.

• Experience working with complex clinical datasets, such as medical imaging, multi-omics, longitudinal patient records, or data from clinical studies and multi-institutional cohorts.

• Familiarity with weakly supervised learning, multiple-instance learning, survival analysis, or related methods.

• Experience with self-supervised representation learning or foundation models.

• Familiarity with dataset shift and variation across sites, devices, scanners, or acquisition protocols.

• Exposure to machine learning in regulated healthcare environments, including SaMD, FDA 510(k)/De Novo, design controls, or CLIA/LDT validation.

• Research experience through publications, conference presentations, internships, or academic projects.

• Familiarity with cloud-based ML development, including distributed training, workflow orchestration, experiment tracking, or reproducible pipelines.

Originally posted on Himalayas