Senior Machine Learning Engineer

About the role

The An-Cockrell Lab at the University of Vermont seeks a subject matter expert in applied machine learning to lead the technical development of AI/ML models for the lab’s ARPA-H-funded research portfolio. Working independently with general direction from the Principal Investigator, this position designs, builds, and deploys machine learning models in a secure Amazon Web Services (AWS) environment using modern tools such as SageMaker. The role also sets standards and practices for other lab staff, oversees development of models that use electronic health record (EHR) and medical sensor data to predict the need for life‑saving interventions, and serves as a technical lead on the CIRCLE project—a multi‑year effort to build a digital twin around the Innate Immune Response Agent‑Based Model (IIRABM). This includes designing multi‑organ compartment and pharmacokinetic/pharmacodynamic (PK/PD) extensions, applying active learning methods (Genetic Algorithms, Deep Reinforcement Learning) to calibrate the model and develop curated drug regimens, and applying large language model (LLM) fine‑tuning and prompt engineering to medical applications. The position trains, guides, and reviews junior technical staff, acts as a primary technical contact with ARPA‑H program officers and external collaborators, and contributes technical content, budgets, and computing plans to future grant proposals. The position is based in Burlington, Vermont, and is eligible for full‑time telework per university policy.

Responsibilities

• Design, build, and deploy machine learning models in a secure AWS environment using SageMaker.
• Set standards and practices for lab staff working in the AWS environment.
• Oversee continued development of models that use EHR and medical sensor data to predict life‑saving interventions.
• Serve as technical lead on the CIRCLE project, designing multi‑organ compartment and PK/PD extensions.
• Apply active learning methods, including Genetic Algorithms and Deep Reinforcement Learning, to calibrate the model and develop curated drug regimens.
• Apply large language model (LL