Principal Data Scientist in London

RightShip is the world's biggest third party maritime due diligence organization, providing expertise in global safety, sustainability and social responsibility best practices. We bring together years of industry expertise with the output from analytics and large data sets to provide our safety and environmental scoring systems, recommendations and consultancy services. Using leading data and technology, we aim to set new benchmarks in environmental protection.

We support global initiatives and action influencing practical and impactful change, enabling "win-win" for business and the environment.

We offer a place where you know you are contributing to an organization that is constantly working to ensure ships are as safe as possible so that crew and cargo are protected. We are passionate about maritime efficiency, safety and sustainability practices. We offer generous rewards. Our base salary is competitive, we support employee wellbeing and provide our employees with a Healthy Living Allowance and our annual incentive scheme is competitive. We have some great talent who share their experience and skills to help you on your way and we are committed to professional development to make sure your career keeps growing while you're working with us.

RightShip is an equal opportunity employer, and we champion diversity. Our teams are composed of individuals from different geographies, cultures, religions, ethnicities, races, genders, sexual orientations, abilities, and generations. We believe that a diversity of experiences makes us stronger—as individuals, as communities and as an organization.

RightShip is scaling its use of AI from traditional ML models to advanced LLM-based reasoning systems, to support safer and more sustainable maritime operations. We are looking for a Data Scientist who can bridge both worlds: classical data science and modern AI/LLM governance. This role is responsible for defining how models should be evaluated, monitored and governed across their lifecycle. You will work closely with domain experts, the existing AI team, data engineering and product teams to ensure our AI systems behave consistently, accurately and responsibly. This is not an LLM-engineering or roadmap ownership role. It is a senior position focused on methodological leadership, model quality, and governance across AI and ML systems.

Major Responsibilities

• Develop experimental methodologies for both classical ML and LLM-based systems.
• Standardise how experiments are structured, documented and compared across teams.
• Create reproducible workflows for testing prompts, model versions, embeddings and classical feature models.
• Ensure statistical soundness and methodological rigour in all experiments.
• Own the Evaluation Framework for Model Behaviour: Build and maintain evaluation datasets, test suites and metrics for reasoning accuracy (LLMs), classification performance (traditional models), consistency and drift, hallucination checks, edge-case handling, and domain-alignment.
• Continuously refine the evaluation approach as the AI team iterates on models.
• Translate Domain Logic into Structured Model-Ready Logic: Work with maritime SMEs to understand inspection logic, safety reasoning and operational decision pathways.
• Convert that knowledge into classification schemas, decision flows, label definitions, reasoning templates, and structured input/output rules for LLMs and ML models.
• Support AI Team Experimentation by providing guidelines, quality criteria and methodology for the AI team's test design.
• Review final experiments and highlight gaps, inconsistencies or risks.
• Maintain the benchmark suite the AI team must compare against.
• Conduct Traditional Data Science & ML Work including analysis of structured datasets, feature engineering and exploratory data analysis, building classification or scoring models as needed, and supporting domain teams with insights, prototypes or predictive modelling.
• Build processes for ongoing monitoring of AI behaviour and model drift, investigate anomalies or unexpected outputs, identify failures rooted in data, logic, edge cases or prompt structure, and recommend corrective actions.
• Maintain transparent documentation, audit trails and governance standards aligned with RightShip's risk posture.

Qualifications, Skills & Experience

• Bachelor's or Masters (preferred) degree in Data Science, or relevant field.
• 7+ years in Data Science, Applied ML or AI-focused role.
• Postgraduate studies or AI governance training is advantageous.
• Curious, a critical thinker and a strong sense of logic.
• Superior mathematical ability.
• Strong experience building and evaluating classification and supervised ML models.
• Excellent communication and presentation skills.
• Excellent at establishing standards others can follow confidently.
• Proficiency in Python, SQL and model evaluation frameworks.
• Familiarity with MLflow (or equivalent), Vector DBs or LLM orchestration tools is a plus.
• Ability to influence through expertise rather than authority.

RightShip is an Equal Opportunity Employer and values diversity, enables access and promotes inclusion in our workplace. You must have the right to live and work in this location to apply for this job.