Required Skills & Experience
- 3 years of experience in a Machine Learning Engineering, Data Science or Data Engineering role
- 2 year of experience in implementing machine learning algorithms in a production environment or applying software development lifecycle principles to analytics
- Hands-on experience designing, building, or integrating LLM-based or agentic applications (e.g., LLM-based applications, RAG systems, or tool-using agents) in a
- production or pilot setting, including retrieval over enterprise knowledge, tool/function calling, workflow handoffs, and secure integration with enterprise systems.
Nice to Have Skills & Experience
- Preferred Qualifications (preferred experience, education, certifications)
- Experience building agentic or LLM applications using orchestration frameworks (e.g., LangGraph, LangChain, Semantic Kernel) and Azure Foundry and OpenAI / OpenAI APIs
- Experience engineering AI-enabled applications or services that integrate APIs, backend services, authentication/authorization, enterprise identity, user-facing interfaces, and platform services in a secure production environment.
- Experience making generative/agentic systems safe for regulated use: evaluation harnesses and guardrails (e.g. evaluation harnesses, safety filtering, , permission-aware responses, identity verification, session isolation, and auditable configuration patterns).
- Experience with vector stores / embeddings and RAG pipelines for domain-specific applications
- Experience working with healthcare data (payer or provider) in a HIPAA-regulated environment
- Epic certification or badges (e.g., Cogito, Cognitive Computing Platform, Chronicles, Interconnect)
- Azure certifications (e.g., AI Engineer Associate, AI Fundamentals, Data Science, or Data Engineering)
Job Description
The Agentic AI / ML Engineer builds, deploys, and operates production AI and machine learning
systems for clinical and operational use across the pediatric health system. The role spans both
predictive machine learning (ML) models and generative and agentic AI — LLM-based
applications, RAG, copilots, and workflow agents — and owns the evaluation, guardrail, and
monitoring frameworks that make these systems safe and reliable to run in a regulated
environment. Working closely with data scientists, clinical informaticists, and EHR/application
teams, this role moves solutions from prototype into production and serves as a technical lead
for agentic AI/ML delivery — setting the standards other engineers and data scientists build
against, and leading incident response for production systems. The role also helps define
enterprise AI system patterns for integrating foundation models into production applications,
including tool/function calling, retrieval over enterprise knowledge, permission-aware
responses, system handoffs, conversation/session management, and reusable configuration
patterns for AI-enabled workflows.
This position will pay $65-85/hr