Data Scientist

Job Description

We are building a new Underwriting Analytics function, a team of data scientists embedded within Underwriting and aligned to our product lines. You will be one of the first hires, helping shape how analytics supports underwriting decisions across the group. Working day-to-day with underwriters, portfolio managers and product heads, your challenge is to take real underwriting problems and use analytics and technology to deliver practical solutions.

Responsibilities

Shape the analytical agenda. Partner with underwriters and product heads to identify, frame and prioritise the highest-value opportunities, turning loosely defined business questions into clear analytical plans, measures of success and deliverables.

• Build line-of-business data strategies. Assess the data available for your aligned product lines, identify material gaps and help create feedback loops linking submissions, exposures, underwriting decisions, claims and portfolio outcomes.
• Improve risk selection and portfolio steering. Develop analysis, segmentation, triage approaches and decision-support tools that help underwriting teams assess opportunities, manage referrals, refine appetite and monitor portfolio performance.
• Strengthen distribution and productivity insight. Analyse broker pipelines, submissions, quote-to-bind conversion, not-taken-up business and portfolio opportunities so underwriting effort is focused on the most valuable flow.
• Deliver practical workflow improvements. Work with underwriting teams to identify friction in day-to-day processes and prototype targeted solutions, such as submission triage, data pre-fill, document automation and lightweight analytical applications.
• Develop a deeper view of market-cycle dynamics. Combine rate, terms, attachment points, deductibles, sublimits, wordings and coverage information to identify early changes in trading conditions and support timely underwriting action.
• Take solutions from discovery to adoption. Explore problems with users, develop and test prototypes, gather feedback, measure impact and support adoption so insight becomes part of everyday underwriting decisions rather than a one-off analysis.
• Work in partnership with central data and technology teams. Use approved platforms and standards, collaborate on data engineering and productionisation, and ensure solutions meet governance, security, support and maintenance requirements.
• Communicate insight clearly. Present findings, limitations and recommended actions to technical and non-technical stakeholders, adapting the level of detail for underwriters, product leaders and senior management.
• Support the development of the function. Contribute reusable methods, shared standards and examples of good practice, while keeping current with relevant developments in analytics, machine learning and generative AI.

Qualifications

Knowledge and Experience

• Applied analytics capability. Strong hands-on experience using data science or advanced analytics to solve commercial problems, ideally in specialty insurance, reinsurance, underwriting, pricing, claims, exposure management, portfolio analytics or another financial-risk environment.
• Python and SQL. Proficiency in Python and SQL, including preparing, analysing and modelling large, imperfect and varied datasets.
• Statistical modelling and machine learning. A sound grounding in statistical methods and machine learning, with the judgement to choose approaches that are proportionate, explainable and usable in an underwriting context.
• Data and platform awareness. Working knowledge of data pipelines, ETL and modern cloud data platforms, sufficient to define requirements and collaborate effectively with data engineers and technology teams.
• Decision-support products. Experience creating trusted dashboards, management information, analytical applications or workflow tools that put insight directly into users' hands.
• Problem ownership. Comfort working with ambiguity and taking a problem from discovery and data assessment through analysis, prototyping, testing and implementation.
• Stakeholder translation. The ability to understand an underwriting or commercial challenge, translate it into an analytical problem and explain the output clearly in business language.
• Insurance and underwriting understanding. Knowledge of specialty or Lloyd's market insurance concepts, or the ability to build domain knowledge quickly and apply it to risk selection, portfolio management and underwriting workflows.
• Modern analytical techniques. Experience with, or a strong practical understanding of, NLP and generative-AI techniques for document-heavy or analytical workflows, including appropriate consideration of explainability, controls and governance.
• Collaborative delivery. A track record of working across business, data and technology teams to deliver solutions within agreed platforms, standards and governance.

Skills

• Communication \& Presentation Skills (Proficiency Level 2)
• Data Intelligence (Proficiency Level 2)
• Data Cleansing \& Wrangling (Proficiency Level 2)
• Development Lifecycle, Processes and Standards (Proficiency Level 2)
• Stakeholder Management (Proficiency Level 2)
• Technical Reporting (Proficiency Level 1)
• Algorithm and Mathematical Model Design (Proficiency Level 2)
• Data Architecture \& Modelling (Proficiency Level 1)
• Data Integration (Proficiency Level 1)
• Data Visualisation \& Storytelling (Proficiency Level 2)
• Solution Design \& Implementation (Proficiency Level 1)
• KPI/Metrics (Proficiency Level 1)