About the Role:

Our client is a software OEM that develops an AI-powered intelligent automation platform that helps insurance enterprises put AI agents to work across complex business processes - from distribution and underwriting to policy servicing and claims. Our engineering teams build the platform, accelerators and integrations that leading insurers rely on every day.

Must-Have Skills: Engineering:

- Expert TypeScript, with deep knowledge of runtime performance and concurrency.

- Experience architecting large frontend and backend systems (Angular and NestJS preferred) as scalable, resilient services.

- Strong system design: distributed and event-driven systems, idempotency, caching, and failure handling.

- Advanced PostgreSQL: data modelling, query tuning, and zero-downtime migrations.

- Production delivery on Docker and Kubernetes with CI/CD.

- Strong security grounding: OAuth 2.1/OIDC, multi-tenant isolation, and secrets management.

Must-Have Skills: AI & LLM:

- Architected production LLM or agentic systems: orchestration, tool design, and context engineering.

- Designed RAG over enterprise data, with access control, freshness, and grounding.

- Owned an evaluation strategy: offline eval suites, regression gates, and quality and cost metrics.

- Applied guardrails: prompt-injection defences, tool sandboxing, PII handling, and audit trails.

- Able to lead AI-assisted engineering practices on the team (agent workflows, spec-driven development, review standards).

Nice-to-Have Skills:

- Designing production MCP servers: tool schemas, authorization, tenancy, and versioning.

- Multi-agent patterns and agent-to-agent protocols (A2A).

- Online evaluation: A/B and shadow testing of model and prompt changes.

- MLOps and model serving (MLflow, vLLM), and choosing between prompting, RAG, and fine-tuning.

- LLM observability and FinOps: tracing, token budgets, and per-tenant cost attribution.

- Insurance or regulated-industry AI: explainability, model governance, and the EU AI Act.

- Infrastructure as code (Terraform, Helm) and Nx monorepos.

Expected Responsibilities:

- Lead the technical design of complex AI-powered features and platform services, documented in design docs and ADRs.

- Define how agents, MCP servers, models, and core services fit together for your product area.

- Break large, ambiguous problems into well-scoped work that several engineers can deliver in parallel.

- Own the reliability, quality, safety, and cost of AI and core services in production.

- Set the evaluation and release bar for model, prompt, and tool changes.

- Drive root-cause analysis for incidents, including model and data regressions.

- Mentor SDE I/II engineers on both distributed systems and AI engineering, and contribute to hiring.