Description

You will define best practices for AI/LLM experimentation, own evaluation frameworks for model behaviour, and translate domain logic into structured model-ready logic. This role focuses on methodological leadership, model quality, and governance across AI and ML systems to support safer maritime operations.

Requirements

• Expertise in classical ML and modern AI/LLM governance
• Strong experience with experimental design and statistical rigor
• Ability to build evaluation frameworks for accuracy, consistency, drift, and hallucination
• Experience working with domain experts to convert knowledge into structured logic
• Senior-level methodological leadership skills

Responsibilities

• Establish experimental methodologies for classical ML and LLM-based systems
• Standardize experiment structures and documentation across teams
• Build and maintain evaluation datasets, test suites, and metrics for model behavior
• Continuously refine evaluation approaches as models iterate
• Translate maritime domain logic into classification schemas, decision flows, and reasoning templates