Senior Data Scientist, Graph Risk Analytics

Location: London, UK Annual Remuneration : £105,000 - £120,000 and discretionary annual bonus Pay Frequency : Monthly Probationary Period: 180 days Work Arrangement : Hybrid, 3-days per week TL;DR Kharon is seeking a full-time Senior Data Scientist, Graph Risk Analytics based in London.

Responsibilities:

Taking ownership of Kharon's risk propagation and entity analytics system — the engine that evaluates entities in our knowledge graph and surfaces meaningful risk signals to customers Designing and implementing complex analytical logic that translate nuanced geopolitical and financial risk into structured, interpretable outputs Building and maintaining scalable data pipelines and graph-based ML systems that power risk scoring and entity analysis Partnering closely with Research, Product, and Engineering teams to understand evolving risk frameworks and translate them into robust, maintainable systems Iterating on the system's architecture to increase the complexity and precision it can handle — this is not a rip-and-replace, but a thoughtful evolution of a system already in production Exploring how LLMs and other AI techniques can be integrated to handle increasingly complex risk logic A proven interest or experience in global security, financial crimes, sanctions, export controls, or related domains 4+ years of experience as a Data Scientist, with demonstrated ownership of production systems Strong foundation in statistics and the ability to apply quantitative methods to complex, real-world problems Proficiency in Python, SQL, and working with graph databases or knowledge graph structures Experience with graph analysis, graph ML, or network analysis — professional experience is a strong differentiator Comfort working across research and data engineering contexts — this role spans all three Experience with Docker, Kubernetes, and API development Strong product instincts — able to go deep technically while staying aligned to customer and business outcomes Bonus: Experience with LLMs or AI-assisted analytical systems