LLM, RAG & Agentic AI Engineer – Insurance Sector in City of London

Occasional travel to client offices and two trips to London HQ per month.

Role Overview:

Lead the design and delivery of AI-native transformation initiatives for insurance clients, spanning agentic systems, retrieval architectures, semantic layers and decision intelligence. This is a senior, hands-on consulting role combining deep AI engineering expertise with strong client-facing presence, shaping both insurance-specific client outcomes and the firm’s long-term AI engineering capability.

As demand accelerates across claims automation, underwriting decision support, policy servicing, fraud detection, compliance and operational efficiency, the consulting practice is expanding its engineering capability across agentic systems, retrieval, ontologies and AI-enabled execution within regulated insurance environments.

The Consulting Engineer is a hands-on AI systems builder who combines engineering depth with commercial and product thinking to design, build and deploy LLM- and agent-driven solutions for insurers, brokers and value chain partners. You will work directly with senior insurance stakeholders (Claims, Underwriting, Operations, IT, Risk, Compliance, Actuarial) and alongside consulting and orchestration roles, translating complex insurance problems into safe, reliable and auditable AI solutions.

Key Accountabilities:

• Client-Facing AI Engineering & Agentic System Design (Insurance-Focused): You will design and deliver production-grade AI systems for insurance clients, including:

• LLM-powered applications for claims handling, underwriting support, policy servicing, document processing and customer operations.
• Multi-agent architectures for insurance workflows, including triage, decision support, escalation, delegation and human-in-the-loop controls.
• Retrieval and vector-based systems over policy wordings, endorsements, claims files, loss runs, underwriting guidelines and regulatory documentation.
• Semantic layers, ontologies and knowledge models aligned to insurance data structures, coverage logic and risk taxonomies.
• Integrations with core insurance platforms (claims systems, PAS, underwriting workbenches), data warehouses and third-party providers.
• Prompt engineering at scale with regulatory guardrails, explainability, traceability and auditability.
• Safety constraints for hallucination control, coverage interpretation accuracy and customer-facing use cases.

• Translate ambiguous insurance challenges into clear, feasible AI solutions.
• Assess client data maturity, policy document quality, legacy platforms and security constraints.
• Shape use cases across claims leakage reduction, underwriting efficiency, fraud detection and compliance automation.
• Work directly with insurance SMEs to surface edge cases, exceptions, regulatory nuances and operational realities.
• Produce clear, concise technical artefacts suitable for regulated, risk-aware client audiences.

• Implementing evaluation frameworks for accuracy, coverage interpretation, decision consistency and bias.
• Designing monitoring, logging and tracing suitable for regulated insurance environments.
• Applying governance, risk and compliance principles (e.g., audit trails, explainability, access controls).
• Supporting controlled releases and operational handover into insurer IT and operations teams.
• Ensuring reliability, reproducibility, performance and cost discipline at insurance scale.

• Build reusable insurance-specific accelerators, agent patterns and reference architectures.
• Contribute to internal playbooks covering claims, underwriting, policy servicing and compliance use cases.
• Share emerging research, frameworks and AI trends relevant to the insurance sector.
• Influence delivery methodology, technical standards and agentic design patterns for regulated industries.

Experience & Skills:

This is a hands-on consulting engineering role. Candidates should bring:

• Experience in software engineering, AI engineering or applied data engineering.
• Strong hands-on experience with LLMs, embeddings, RAG pipelines and vector databases.
• Experience designing or implementing multi-agent systems or tool-calling frameworks.
• Strong Python skills with experience building production-grade, regulated systems.
• Experience with at least one major cloud AI ecosystem (Azure/OpenAI, GCP/Vertex, AWS, Anthropic).
• Familiarity with semantic modelling, ontologies or knowledge graph concepts, ideally applied to complex domains.
• Proven ability to rapidly prototype and validate solutions with business stakeholders.
• Experience working directly with clients in consulting, professional services or regulated enterprise environments.
• Insurance domain experience (claims, underwriting, policy, risk, compliance or adjacent systems) strongly preferred.