[Hiring] Senior Agentic AI Enablement Engineer @Crum & Forster

Role Description

Crum & Forster is building an enterprise agentic AI capability across underwriting, claims, legal and compliance, and corporate operations, with engineering teams. We are seeking a Senior Agentic AI Enablement Engineer to design, build, and ship agentic AI solutions hands-on. You will turn validated solution designs into working agents, build the reference implementations our engineering pods build on, and work embedded alongside our AI execution pods to accelerate production delivery.

This is a hands-on engineering role. Most of your time will be spent writing production-quality code: multi-agent orchestration, retrieval pipelines, tool integrations, and evaluation harnesses on AWS. You will work closely with the Principal AI Engineer, who sets architecture and technical standards, and you will be the US-based build anchor for our agentic platform.

Why This Role: You will build the agentic AI systems at the heart of C&F’s AI transformation, hands-on and on a modern AWS stack, while shaping how our engineering teams build. The role offers a clear growth path toward Principal AI Engineer as our AI capability scales.

What you will do:

• Agentic Solution Engineering:
  • Build multi-agent workflows across planner, retrieval, tool-calling, and validation steps, including context engineering, agent memory, state management, and escalation paths.
  • Implement autonomy boundaries, human-in-the-loop checkpoints, deterministic guardrails, and audit trails in code.
  • Develop and integrate tools and services through Model Context Protocol (MCP) and enterprise APIs, including policy administration, claims, and document systems.
• Retrieval & Grounding:
  • Build retrieval pipelines over unstructured submissions, claims files, and financial records: document parsing, chunking, embeddings, hybrid search, and reranking.
  • Tune retrieval quality and grounding using measurable metrics such as hit rate, MRR, and citation accuracy.
• Prototyping & Reference Implementations:
  • Rapidly prototype agents on Amazon Bedrock and Bedrock AgentCore to prove feasibility before production build.
  • Build reusable reference implementations, SDK components, and patterns that the execution pods adopt and extend.
  • Harden successful prototypes into production-ready components to the engineering standards set by the Principal AI Engineer.
• Evaluation, Observability & Reliability:
  • Build evaluation harnesses with golden datasets, LLM-as-judge scoring, and regression gates that run in CI/CD before release.
  • Instrument agents with tracing, observability, latency and cost monitoring, and work with the reliability engineering pod on production quality.
• Delivery Partnership:
  • Work embedded alongside the AI execution pods: pair on hard problems, review code, and unblock delivery.
  • Contribute to technical design reviews and architecture decisions with the Principal AI Engineer.
  • Mentor engineers on agentic engineering practices across US and India teams.

Qualifications

• Bachelor’s degree in Computer Science, Engineering, or a related field or equivalent experience.
• 8+ years of software engineering experience, including 3+ years building Generative AI or agentic AI solutions, with at least one in production.
• Strong Python engineering skills and experience writing production-quality, tested code.
• Hands-on experience with LLM orchestration frameworks such as LangGraph, LangChain, or equivalent, including tool calling and multi-step agent workflows.
• Hands-on experience building retrieval-augmented generation pipelines with vector databases, embeddings, and reranking.
• Experience with AWS, including serverless and containerized services.
• Experience implementing LLM evaluation and testing approaches.
• Solid grounding in API design, CI/CD, and secure development practices.

Preferred Qualifications

• Experience with Amazon Bedrock, Bedrock AgentCore, Bedrock Guardrails, or Bedrock Knowledge Bases.
• Experience building or deploying MCP servers and tool integrations.
• Experience with agent observability and tracing tools such as LangSmith or OpenTelemetry.
• Experience with TypeScript or another modern language in addition to Python.
• Experience in Property & Casualty insurance or another regulated industry.
• Experience working with distributed teams across US and India time zones.

What Success Looks Like

• Working agentic prototypes that prove feasibility quickly and move into production build.
• Reference implementations and reusable components adopted across the AI.
• Agents released with evaluation gates, observability, and controls built in from the start.
• Measurably faster delivery from the engineering teams you work alongside.

Benefits

• Competitive compensation package
• Generous 401K employer match
• Employee Stock Purchase plan with employer matching
• Generous Paid Time Off
• Excellent benefits that go beyond health, dental & vision, focused on your whole family’s wellness
• Support for career development, including tuition reimbursement, industry-related certifications, and professional training
• A dynamic, ambitious, fun and exciting work environment
• Encouragement of social and community responsibility through matching donation programs and volunteer opportunities