Lead AI System Architect

This is a fully remote position, open to applicants in Ireland.

📋 Description

• Take charge of the architecture for EIS's agentic platform: encompassing agent orchestration, MCP-native tool ecosystems, agent memory (short-term, long-term, semantic), planning, and reusable tool/function calling patterns across various product domains.

• Facilitate and support domain teams focusing on vertical insurance agents and the horizontal capabilities (RAG, retrieval, instructional flows) they develop.

• Establish and uphold levels of autonomy — assistive, semi-autonomous, autonomous — incorporating explicit human-in-the-loop checkpoints, escalation routes, and reversibility for high-stakes actions within regulated workflows.

• Lead the MCP strategy: determining which capabilities EIS presents as MCP servers to both internal and partner agents, how our agents utilize external MCP tools, and the tool registry, schemas, and versioning necessary for scalability.

• Uphold the multiple stack approach as a core capability: Typescript and Java. Guide teams in selecting the appropriate stack for each agent while ensuring alignment through shared configuration artifacts, prompt management, and evaluation tools.

• Oversee Architecture Decision Records (ADRs) for agentic capabilities; collaborate with Platform, Security/InfoSec, and DevOps to ensure agents are observable, testable, sandboxed, and compliant by default.

• Propel AI DevOps for agents: including trace capture and replay, evaluation harnesses (task success, tool-use correctness, regression), prompt and model versioning, cost and latency budgets per agent, and progressive rollout strategies.

• Establish safe-AI standards for agentic systems: developing defenses against prompt injection and tool-poisoning, action allow-lists, blast-radius controls, PII management, data residency, and bias mitigation. Prioritize agent safety as a fundamental architectural concern.

• Convert insurance use cases into actionable production agent designs in collaboration with product strategists and domain architects; provide technical leadership and mentorship; effectively communicate agentic trade-offs (autonomy, reliability, cost, safety) to executives, customers, and engineers.

⛳️ Requirements

• Demonstrated success in designing and delivering agentic systems in production — focusing on real-world applications, not demos or prototypes — featuring significant autonomy and multi-step tool usage.

• Strong background in systems design: experience with data-intensive, distributed, and latency-sensitive environments in production.

• Extensive, hands-on knowledge of agent patterns: orchestration, planning, ReAct-style and graph-based agents, agent memory, tool/function calling, MCP, and structured outputs. Keen insight into when to employ an agent versus a deterministic workflow, tracking the latest developments and translating them into actionable roadmaps.

• Proficient in the Java/Spring ecosystem.

• Experienced with Typescript and Python for AI (LangChain, LangGraph, or a similar agent framework) — production experience is essential. Comfortable working in both technology stacks.

• Practical experience with vector databases including embedding models, hybrid search, re-ranking, and retrieval evaluation.

• Proficient in agent evaluation and observability: including traces, replays, evaluation harnesses, guardrails, and cost/latency telemetry. Familiar with AI configuration-as-code.

• Experience in deploying AI services on cloud platforms (AWS, Azure, GCP) within regulated enterprise settings — including security reviews, data residency, and audit trails.

• Familiarity with the insurance, financial services, or another regulated sector is advantageous.

• Strong architectural judgment — practical considerations regarding build versus buy, vendor versus in-house solutions, agent versus deterministic workflows, model selection, and overall cost of ownership.

• Exceptional written and verbal communication skills; capable of making agentic trade-offs understandable to non-AI audiences.

• Advanced degree in Computer Science, AI/ML, or a related field — or equivalent practical experience.

🏝️ Benefits

• Collaborate with top-tier talent and exceptional colleagues who are experts in their fields. Work in a Scaled Agile environment with diverse, multicultural, and cross-functional teams.

• Join a global and innovative software product company that develops world-class Enterprise InsurTech Products powered by cutting-edge technologies (microservices, reactive, cloud, continuous delivery).

• Utilize the latest Apple MacBooks for your work.

• Enjoy the freedom to shape your career path through development programs and exciting global mobility opportunities (we foster a global culture).

• **Incentives and Benefits:** Allowances:

• Homeworking Equipment

• Mobile phone/internet coverage

**Benefits:**

• Health Insurance for you and your dependents — eligibility begins immediately upon joining EIS.

• Optional sports coverage within the health insurance package.

• Pension contributions based on a Group Pension Scheme.