Senior Data Scientist

Senior Data Scientist (Permanent) — London (Hybrid, 3+ days in office)

Gravitas is partnering with a leading Lloyd’s market insurer to hire a Senior Data Scientist into their Data Science & Analytics function.

Compensation: £75,000–£95,000 base + 20% bonus (plus benefits).

Role overview

Reporting to the Data Science Manager, you’ll strengthen the firm’s data science capability by delivering models and actionable insights that improve underwriting profitability and unlock automation and efficiency across teams.

This is a true end-to-end role: you’ll own projects from problem framing through development and deployment, and remain accountable for models in life—monitoring performance and drift and deciding when retraining or retirement is required. You’ll design with a road-to-production mindset from day one, with demonstrable experience of personally taking models into operational use.

You’ll also work continuously with commercial underwriters, translating underwriting requirements into data science solutions, building confidence in outputs, and spending time with underwriting teams to understand how each class operates. Close collaboration with Actuarial is expected from the outset.

Key responsibilities

Delivery of data science products

• Lead data science projects end to end: problem framing, data prep, modelling, deployment, and ongoing production monitoring.
• Partner with actuarial colleagues to surface insights that drive performance (e.g., reserving).
• Apply data science techniques to automate manual processes across the business.
• Use generative AI to enrich insight and unlock roadmap opportunities, deploying and maintaining these solutions via robust MLOps patterns.
• Research, assess and integrate external data sources for quality, value and fitness for use.
• Address data quality issues constraining modelling (including premium/claims matching for delegated business).
• Support proactive analytics and insight delivery across the business.

Engineering & MLOps standards

• Design, build and maintain ML pipelines in a cloud environment (Azure-based).
• Raise standards across version control, testing, CI/CD, model versioning and reproducibility.
• Own deployed models in life: monitor drift/performance and act before business impact.
• Ensure models are documented and explainable to a regulated-environment standard.

Stakeholder engagement & requirements

• Identify, document, analyse and prioritise requirements across technical and non-technical stakeholders.
• Coordinate with IT/Data Engineering to shape the data foundations these products depend on.
• Produce clear deliverables and communicate findings (and limitations) to non-technical audiences.

Team & capability building

• Coach data scientists and analysts via code review, pairing and technical mentoring.
• Support upskilling in emerging techniques while maintaining clear accountability.
• Contribute to backlog and roadmap, advocating for projects with demonstrable value.

Essential skills & experience

• Strong Python to production standard (OOP, testing, code review).
• Proven experience taking models into production and supporting them in life.
• Strong ML/statistics toolkit (e.g., pandas, NumPy, scikit-learn, statsmodels or equivalent) and sound validation judgement.
• Software engineering fundamentals: version control, branching strategy, code review, automated testing, dependency/environment management.
• Practical MLOps/CI/CD: orchestration, versioning, automated deployment, monitoring, retraining patterns.
• Cloud ML delivery (ideally Azure ML / Azure DevOps; AWS/GCP considered).
• Strong SQL and relational data modelling; comfortable with large datasets.
• Data wrangling of incomplete/inconsistent real-world data (common in insurance).
• Statistical foundations to design experiments, quantify uncertainty and challenge unsupported conclusions.

Desirable

• Lloyd’s/insurance pricing or underwriting experience in a regulated environment; comfort working alongside actuarial methodology.
• Hands-on generative AI / LLM deployments (retrieval patterns, evaluation, cost/latency, observability).
• PySpark / distributed processing.
• Power BI or similar visualisation/reporting.

Package & location

• £75k–£95k base + 20% bonus
• Permanent, full-time
• London (hybrid) — minimum 3 days/week in-office