Essential Responsibilities:
- Develop and evaluate AI-based biomarkers using multimodal data.
- Design, implement, and improve machine-learning models to predict patient outcomes and treatment response.
- Contribute to the end-to-end model development lifecycle, including data preparation, training, evaluation, and validation.
- Support the productionalization, launch, and monitoring of machine-learning models in collaboration with platform and product teams.
- Conduct research and experimentation to improve model performance, robustness, generalizability, and interpretability.
- Collaborate with biostatistics, clinical, and product partners to translate clinical questions into machine-learning solutions.
- Contribute to scientific publications and conference submissions alongside the broader research team.
Experience Requirements:
- 2+ years of industry experience using PyTorch or TensorFlow.
- Experience contributing to machine-learning systems deployed or maintained in production environments.
- Ability to clearly communicate complex technical concepts to cross-functional, non-ML collaborators.
Desired:
- Experience working with large-scale image data or computer vision models.
- Familiarity with self-supervised representation learning (e.g., MoCo, DINOv2) and / or vision–language models (VLMs) and multimodal representation learning.
- Interest in healthcare, medical imaging, or applied machine learning in regulated or high-impact domains.
