About this role
About the Role Three hundred fifty million Americans rely on a healthcare system whose decision-making has become slow, costly, and adversarial. Deloitte has an AI-first effort, backed by a $1B committed investment, to rebuild the reasoning models and agentic systems behind healthcare decisions at national scale. This role supports real post-training at scale with substantial GPU compute and training infrastructure, not toy fine-tunes. This is a ground-up rebuild of the decision-making machinery behind American healthcare, at national scale. As a Research Engineer, you will contribute to building the next-generation post-training capabilities and small language models for Healthcare AI Data and Analytics.
What You'll Do
- Design, train, evaluate, and deploy post-training language models and small language models (SLMs) for healthcare analytics.
- Develop reasoning models and agentic systems to support faster, fairer, and more reliable healthcare decision-making.
- Collaborate with data engineers and AI infrastructure teams to scale GPU compute and training pipelines.
- Experiment with evaluation metrics, safety, and reliability in healthcare contexts.
What We're Looking For
- Experience in ML research related to NLP, large language models, and post-training methods; strong programming skills (e.g., Python, ML frameworks).
- Background applying ML research to healthcare data and analytics; ability to translate business problems into technical experiments.
- Ability to work with cross-functional teams to deliver production-ready capabilities.
- Strong problem-solving, communication, and collaboration skills.
Nice to Have
- Experience with healthcare AI, data governance, and privacy/compliance considerations.
- Familiarity with large-scale training infrastructure and cloud environments (GPU clusters, storage, orchestration).
Compensation & Benefits
- Salary and benefits information not disclosed in the posting.