The posting
We are seeking a Senior AI/ML Engineer to design, build, deploy, and evolve AI models, agents, and workflow automation for clinical terminology and content operations. This hands-on role combines AI/ML development with production ownership, taking models and agents from experimentation to reliable production use. The ideal candidate has experience with large language models, agent frameworks, retrieval-augmented generation, and the infrastructure and controls required to operate AI systems reliably.
WHAT YOU’LL DO:
- Develop machine learning models, agents, and automation workflows for terminology management, content creation, mapping, and validation — evolving them from experimentation into scalable production systems.
- Build agentic workflows that use LLMs, tools, APIs, knowledge sources, retrieval capabilities, and structured business rules to complete complex tasks.
- Build and maintain retrieval-augmented generation solutions, vector and semantic search capabilities, and prompt and context-management strategies.
- Partner with our data science team to understand, integrate, and productionize their existing agents, and bring your own model and agent development to the team's roadmap.
- Own the deployment, monitoring, troubleshooting, and continuous improvement of AI workflows in production, including root-cause analysis and durable remediation of failures or unexpected outputs.
- Design evaluation, testing, and observability practices for AI systems, and implement controls for auditability, explainability, and human-in-the-loop review in clinically sensitive workflows.
- Develop cloud-based solutions using AWS services such as Amazon Bedrock, SageMaker, and Lambda, applying CI/CD, containerization, automated testing, and secure development practices.
- Work closely with clinical, mapping, product, data science, and engineering partners to translate workflows into practical solutions — and help define where AI automation is appropriate, where deterministic logic is required, and where human review must remain.
WHAT YOU’LL NEED:
- 5+ years across AI/ML engineering, data science, machine learning engineering, or related disciplines, with a foundation in applied machine learning.
- Hands-on experience building agents and agentic workflows, including orchestration and tool or function calling.
- Hands-on experience building RAG solutions, including embeddings, vector databases, semantic search, and context engineering.
- Hands-on MLOps experience taking models and agents into production — deployment, versioning, monitoring, and CI/CD across multiple environments.
- Strong Python proficiency and experience developing maintainable services, APIs, pipelines, or workflow automation, plus working knowledge of SQL and relational databases such as PostgreSQL.
- Experience with cloud-based AI infrastructure, preferably AWS and Amazon Bedrock.
- Strong troubleshooting and root-cause analysis skills, and the ability to partner with domain experts and convert ambiguous workflow needs into scalable technical solutions.
- Clear written and verbal communication in cross-functional environments.
PREFERRED QUALIFICATIONS:
- LangChain or LangGraph, LlamaIndex, OpenSearch, vector databases, or evaluation frameworks.
- Multi-agent or tool-using workflows, including state management, memory, routing, and failure recovery.
- Testing and evaluation approaches for non-deterministic AI systems.
- Healthcare technology, clinical terminology, clinical data normalization, mapping workflows, or regulated data environments.
- Familiarity with healthcare data standards such as knowledge graphs, FHIR, SNOMED CT, LOINC, RxNorm, ICD-10, or CPT.
- AI solutions incorporating human review, auditability, explainability, and quality governance.
