About the Role
Insurance isn't the first industry most engineers think of when they imagine cutting-edge AI work. That's exactly why this role is interesting. CFC's Data & AI unit is building production agentic systems that automate complex underwriting decisions - not chatbots, not copilots bolted onto legacy workflows, but autonomous multi-step AI agents that reason over unstructured data, assess risk, and drive real business outcomes. The team uses frameworks like LangChain to orchestrate LLM-driven services that sit at the heart of how the business operates.
Responsibilities
•Design, develop, and maintain business-critical AI agent services
•Build tailored agent workflows and services from business requirements, using LangChain and related frameworks with reliable patterns for LLM-driven decision-making in production
•Integrate tests and validation to improve AI agents through evals and monitoring
•Work with software engineers and architects to lead system design and architectural decisions
•Translate technical specifications into clean, testable, and scalable production code
•Work closely with cross-functional teams - engineers, data scientists, product managers - to deliver features on time and to a high standard
•Write unit and integration tests to maintain reliability and service correctness
•Monitor, troubleshoot, and continuously improve production services
•Produce clear, structured documentation for systems, architecture, and processes
•Mentor junior engineers through code reviews, best-practice guidance, and knowledge sharing
Requirements
•Significant experience building production-grade AI agents or LLM-powered services
•Strong Python development experience, ideally with 6+ years of professional software engineering experience
•A track record of writing clean, maintainable and high-quality production code
•Experience supporting business-critical systems in live production environments
•Strong understanding of asynchronous programming, Docker, containerised deployments and modern service architectures
•Experience designing distributed systems, asynchronous microservices and event-driven architectures
•Confidence leading system design discussions and making pragmatic architectural trade-offs
•Strong cloud experience, ideally within Azure
•Experience deploying, monitoring and maintaining ML or LLM models in production
•Familiarity with MLOps, LLMOps, evals, monitoring and lifecycle management
•Hands-on experience orchestrating LLM workflows using frameworks such as LangChain
•An interest in how agentic software development can improve engineering delivery without compromising code quality
•Strong communication skills and the ability to collaborate effectively in remote or asynchronous environments
•An ownership mindset, with the ability to work independently and contribute effectively to shared codebases
Conditions
No conditions stated in the posting.
