About the Role We're looking for a Data Scientist to build intelligent AI agents and retrieval systems that power our clinical platform. You'll design multi-step reasoning systems, agent architectures, and the underlying knowledge infrastructure that enables AI to autonomously navigate complex clinical workflows.
What You'll Do - Design and build AI agents capable of multi-step clinical reasoning and tool use
- Develop multi-agent systems that coordinate specialized AI components for complex analysis
- Build intelligent retrieval and RAG systems that ground agent actions in relevant patient data
- Create agent orchestration frameworks that handle diverse clinical workflows
- Implement document understanding pipelines for structured extraction from clinical documents
- Optimize agent performance, reliability, and cost efficiency
Qualifications Required - Strong foundation in deep learning frameworks (PyTorch) and LLM architectures
- Experience building LLM-powered agents or multi-step reasoning systems
- Familiarity with agent frameworks and tool-use patterns
- Experience with retrieval systems, embeddings, and vector databases
- Proficiency in Python and modern ML development tools
- Ability to read and implement research papers
- Excellent communication skills
\
Preferred - Master's degree in Computer Science, Machine Learning, or related quantitative field (PhD preferred)
- Public in ML/AI/NLP venues (NeurIPS, ICML, ICLR, ACL, EMNLP)
- Experience with multi-agent coordination or orchestration systems
- Background in healthcare data or clinical workflows
- Experience with reinforcement learning for agent optimization (PPO, etc.)
- Familiarity with document processing and information extraction

Boston, MA, United States of America
Click apply
JS26489_25303_158E8067C388A909E04C7C46889D4C72
1/27/2026 8:13:23 AM
We strongly recommend that you should never provide your bank account details to an advertiser during the job application process. Should you receive a request of this nature
please contact support giving the advertiser's name and job reference.