Senior Applied & Agentic AI Engineer

Senior Applied & Agentic AI Engineer Job Responsibilities

• Lead the architecture and delivery of enterprise‑grade LLM and agentic AI systems that transform claims, risk, and operational workflows.
• Define technical strategy for retrieval‑augmented generation (RAG), multi‑agent orchestration, and autonomous workflow automation.
• Design and implement advanced agentic systems capable of planning, reasoning, tool selection, execution, reflection, and recovery.
• Architect stateful, memory‑aware AI systems that manage long‑running claims processes across multiple touchpoints.
• Build multi‑agent collaboration models that coordinate coverage analysis, document validation, fraud signals, compliance checks, and decision support.
• Establish orchestration frameworks that manage task routing, context persistence, structured outputs, and failure handling.
• Design secure tool integration layers connecting agents to claims systems, policy platforms, data warehouses, document repositories, and external data services.
• Implement deterministic guardrails, schema validation, and output verification pipelines to reduce hallucination and execution risk.
• Lead development of document intelligence systems leveraging LLMs for summarization, entity extraction, discrepancy detection, and structured data reconstruction.
• Define prompt engineering standards and reusable reasoning templates for consistent, domain‑aware outputs.
• Oversee embedding strategies, vector indexing architecture, retrieval optimization, and knowledge grounding approaches.
• Design evaluation frameworks to measure reasoning depth, workflow completion accuracy, hallucination rates, latency, and cost efficiency.
• Implement observability layers that track agent decisions, tool usage, retrieval effectiveness, and drift across models and prompts.
• Drive optimization strategies for token efficiency, caching, batching, and inference scaling.
• Ensure compliance with Responsible AI principles, enterprise governance standards, audit requirements, and regulatory constraints.
• Partner with enterprise architecture, cybersecurity, and data governance teams to define secure deployment patterns.
• Mentor engineers on LLM orchestration patterns, workflow decomposition, and safe agent design.
• Translate executive‑level business objectives into scalable AI platform capabilities.
• Lead proof‑of‑concepts through full production deployment with measurable ROI outcomes.
• Continuously evaluate emerging foundation models, orchestration frameworks, and agent tooling for enterprise readiness.

• Bachelor's or Master's degree in Computer Science, Artificial Intelligence, Engineering, or related discipline.
• 7-10+ years of experience in AI engineering, machine learning systems, or distributed software architecture.
• 3-5+ years designing and deploying LLM‑powered systems in production environments.
• Demonstrated experience architecting full agentic AI systems with planning, reflection, memory, and tool execution components.
• Deep expertise in RAG architectures, embedding strategies, vector databases, and retrieval optimization.
• Strong experience designing multi‑agent orchestration frameworks and workflow engines.
• Advanced proficiency in Python and enterprise API integration patterns.
• Experience building secure, scalable microservices in cloud‑native environments.
• Strong understanding of distributed systems, event‑driven architectures, and system reliability principles.
• Experience implementing structured output enforcement, guardrails, and audit logging mechanisms.
• Demonstrated ability to design evaluation and benchmarking frameworks for LLM and agent reliability.
• Experience operating in regulated industries such as insurance, financial services, or healthcare preferred.
• Proven leadership in technical design reviews, architecture governance, and cross‑functional collaboration.
• Strong ability to balance innovation with enterprise risk management and operational stability.

Sedgwick is an Equal Opportunity Employer and a Drug‑Free Workplace.