Labcorp Genetics is seeking a Lead AI/ML Scientist to develop a modeling platform that supports the clinical interpretation of genetic testing results for hundreds of thousands of patients annually, leveraging clinical genomic expertise, massive genetics and lab data, and AI/ML technologies to improve diagnosis, clinical care, and treatment options.
Requirements
- Proficiency in Python and machine learning libraries such as PyTorch.
- Track record of deploying machine learning models in production on cloud platforms.
- Familiarity with regulatory, ethical, and data considerations in healthcare AI.
- Hands-on experience in Bayesian statistical modeling and programing with Pyro.
- Experience with containerization of pipelines with Docker and orchestration with Kubernetes.
- Previous work in genetics, functional genomics, or clinical research.
- Track record of impactful publications and patents.
Responsibilities
- Lead the design, development, and evaluation of AI/ML algorithms over genetic and clinical data.
- Assess, test, and integrate published state-of-the-art models.
- Integrate new public and proprietary data.
- Collaborate with machine learning engineers to release models into production.
- Write clean, maintainable, and well-documented code following best practices for version control, testing, and scalability.
- Ensure compliance with data privacy and ethical guidelines throughout the model lifecycle.
- Stay up-to-date with the latest advances in AI/ML research and technologies.
Other
- Applicants who live within 35 miles of either the Burlington, NC or Durham, NC location will follow a hybrid schedule.
- This schedule includes a minimum of three in office days per week at an assigned location, either Burlington or Durham, supporting both collaboration and flexibility.
- Drive interdisciplinary collaborations both internally and externally to ensure clinical and translational relevance.
- Mentor junior scientists and contribute to a culture of creativity and scientific rigor.
- Excellent communication skills for presenting research findings to diverse audiences.
