Senior AI Engineer - Agentic Engineering

We're looking for a Senior AI Engineer - Agentic Engineering to join our team in Nicosia, Cyprus, in a hybrid working mode. In this role, you will build and evolve the software powering AI and agentic systems for reinsurance, advancing capabilities across Claims, Treaty and Facultative lines. You will develop robust tools, services and workflows that Large Language Model (LLM)-based agents use, ensuring production readiness, scalability and reliability. This position allows you to work at the intersection of AI innovation, enterprise platforms and real-world business impact.

Responsibilities

• Maintain, refactor and extend AI applications and supporting services in production
• Build agentic-friendly software components and APIs leveraged by LLM-based agents
• Design AI workflows that integrate models, prompts, tools, enterprise data and business logic
• Validate model behavior and outputs from functional and business perspectives
• Design and maintain evaluation frameworks for AI and LLM systems, including automated and human-in-the-loop evaluations
• Implement quality checks and grounding mechanisms for agent behaviors
• Ensure system performance objectives, including reliability, scalability, latency optimization and cost efficiency
• Collaborate on bringing new agentic solutions into production environments
• Utilize agentic coding harnesses to guide AI agents, review code outputs and maintain standards

Requirements

• Strong software engineering skills in a modern language (Python, TypeScript or Java/Kotlin)
• Experience with clean code principles, automated testing and CI/CD pipelines
• Expertise in API/service/integration engineering and cloud-native development
• Familiarity with LLM application design patterns, such as retrieval-augmented generation and tool/function calling
• Experience working with agentic coding harnesses (e.g., Claude Code) and Linux/cloud environments
• Ability to improve and maintain existing codebases effectively
• Exposure to data science and evaluation methods for AI/ML models
• Fluency in English

Nice to have

• Hands-on experience in agentic or LLM application development (including agent loops, evaluations and custom tool design)
• Familiarity with Palantir Foundry/AIP
• Knowledge of insurance or reinsurance domains
• Experience with cloud AI services (e.g., Azure AI, AWS AI services)