Senior Machine Learning Engineer

Transcarent is the One Place for Health and Care, bringing together medical, pharmacy, and point solutions with a generative AI-powered health and care platform. The Senior ML Engineer will design, build, tune, and evaluate multi-agentic systems that guide users through complex conversations, focusing on agent behavior and model selection.

Responsibilities:

• Design and orchestrate multi-agentic workflows
• Own context engineering for production agents, including system design, safety rules, context injection, and clarifying question strategies
• Design tools and function-calling interfaces, so agents take reliable, well-structured actions
• Build and tune retrieval (RAG) pipelines—embeddings, vector search, filtering, query rewriting, and relevance tuning
• Select and optimize models across providers, balance quality, latency, determinism, and cost
• Design agent memory and context management for coherent multi-turn behavior
• Build safety and guardrail layers for input filtering, scope and safety checks, and graceful handling of edge cases
• Own LLM evaluation, offline eval suites, graders/LLM-as-judge, test sets and personas, metrics, and quality gates
• Collaborate with cross-functional stakeholders on requirements, project execution and status tracking
• Meta technical responsibility: Document high-fidelity technical designs, establish alignment on solutions within broader engineering team

Requirements:

• Bachelor's or master's degree in data science, Machine Learning Engineering, or a related technical field, or equivalent practical experience
• 5+ years of professional Data Science/ML engineering experience
• Strong applied experience building LLM-powered agents in production — shipped, multi-turn agentic systems, not just prompt experiments
• Hands-on expertise with agent orchestration frameworks — stateful graphs, tool use, and conditional routing
• Deep understanding of context engineering and tool / function-calling design for reliable agent behavior
• Practical RAG experience — embeddings, vector search, and retrieval-quality tuning
• Fluency with LLM model selection and tuning across providers, including reasoning models and their trade-offs
• Experience designing LLM evaluation — offline eval, graders, test sets, metrics, and quality gates
• Comfort with agent observability and tracing to diagnose and improve behavior
• Strong Python skills as applied to ML/agent work
• Experience with agent memory systems
• Experience with LangChain suite
• Experience building safety guardrails for high-stakes domains (clinical, financial, legal)
• Experience optimizing LLM latency, cost, and reliability at scale
• Experience with building and working with MCPs and loop engineering
• Prompt optimization techniques such as GEPA
• Working with sensitive data in regulated environment