Overview Join St. Jude's Office of Data Science as an Agentic AI Scientist to design, build, and evaluate agent-based AI systems that support research and operations. You will lead complex AI projects, design agent architectures, and ensure secure, auditable deployment within HIPAA-aligned patterns. You will translate emerging AI capabilities into practical, high-impact solutions while guiding staff and contributing to technical reporting. This role offers the chance to shape how data science accelerates pediatric biomedical discoveries at a world-leading hospital.
Responsibilities- Lead agentic-AI research and engineering projects with moderate supervision
- Design and orchestrate agent systems (hierarchies, tool use, handoffs, checkpointing, human-in-the-loop) for ODS data and tools
- Set up and maintain MCP servers/clients and A2A-interoperable agents connected to St. Jude assets under secure patterns
- Instrument agents with tracing, evaluations, and cost/latency budgets; report results
- Draft manuscripts and technical reports; guide staff in agentic-AI techniques
- Mentor others in agentic AI best practices and standards
Key requirements- Bachelor's in Computer Science, Bioinformatics, Biomedical Engineering, or related field with 6+ years of relevant experience or Master's with 4+ years or entry-level PhD with no work experience
- Publication benchmark: 1-2 first-author papers with journal impact factor >5 or equivalent open-source contributions
- Preferred: prior applied-ML or agentic-AI research; shipping of LLM-integrated systems; healthcare/biomedical domain experience
- collaborative mindset
- mentoring and guiding others
- strong communication and writing
- Agent frameworks (LangGraph, OpenAI Agents SDK, Google ADK, CrewAI, Microsoft Agent Framework, Pydantic AI, LlamaIndex, Strands, Haystack, DSPy, Agno)
- Interoperability and MCP servers/clients; JSON-RPC with OAuth 2.1; A2A interoperability
- Orchestration and control loops; multi-agent topologies; checkpointing and human-in-the-loop
