Senior Data Scientist

About the role

Relay is fundamentally reshaping how goods move in an online era. The role requires you to own the core modelling and production systems that power Relay's digital twin, ensuring that strategic simulations remain accurate and actionable as the network scales. You will dissect high-stakes business questions into structured modelling problems and design solutions that balance statistical rigour with operational reality. A significant portion of your work will involve evolving existing cost and flow components while introducing new capabilities for emerging services and operating models. You will act as the primary technical liaison between advanced data methods and the commercial decisions made by finance and strategy teams. This position demands a builder's mindset, where you translate ambiguous requirements into reliable models and ship components that other teams depend on daily. Ultimately, you will help define what is possible within the network by creating tools that make the right trade-offs between accuracy, speed, and interpretability.

Key facts

What you'll do

Identify and own the core components of Relay's digital twin, evolving models that simulate unit economics, capacity, and network behaviour end-to-end from collection through last-mile delivery.

Upgrade cost engines when new first-mile operating models roll out, ensuring that every new service type, pricing option, or routing choice is represented with granular accuracy.

Analyse thousands of data points per parcel to surface new metrics, build new flows, and harden the parts of the system that matter most for high-impact strategic decisions.

Determine what belongs inside the living digital twin versus what should exist as a standalone strategic model, and select the appropriate tool - classical statistical modelling, financial modelling, simulation, or ML - based on the question at hand.

Partner closely with Strategic Finance, Commercial Finance, and FP&A to translate their pricing, margin, and projection questions into dynamic modelling problems they can explore without manual rebuilds.

Design and maintain production-grade Python and SQL pipelines that feed the digital twin, taking responsibility for data quality, modularity, and maintainability within a real-world codebase.

Collaborate with a data squad of engineers, analysts, and data scientists to prioritise roadmap items, ship new model components, and iterate based on feedback from finance and commercial stakeholders.

Write tested, readable code and design modular APIs so that models and simulations remain understandable and extensible for colleagues joining the project in the future.

Apply financial modelling, forecasting, and simulation experience to scenarios that directly influence commercial strategy, ensuring your work leads to actionable recommendations rather than isolated analysis.

Communicate clearly with non-technical stakeholders, bridging the gap between sophisticated modelling outputs and practical business decisions across the organisation.

Requirements

You hold a PhD or Master's degree in a quantitative field such as computer science, mathematics, or operations research, with advanced training in modelling complex systems.

You bring more than eight years of professional experience in data science, analytics, or modelling roles, with a proven track record of delivering production-grade analytical solutions.

You are fluent in Python and SQL, capable of performing your own data engineering and modelling without reliance on separate data engineering support for core tasks.

You have deep experience in financial modelling, forecasting, or simulation, ideally in contexts where model outputs have directly influenced commercial, pricing, or strategic decisions.

You understand software engineering best practices and apply them rigorously to modelling work, including testing, versioning, and maintainable code design within a production environment.

You are comfortable working with APIs and front-end interfaces enough to collaborate effectively with frontend engineers and product teams on the digital twin stack.

You have a strong grasp of statistical methods, ML techniques, and when each approach materially improves accuracy, reliability, or interpretability in network-level decision making.

You are a systems thinker who can break down complex problems into components, challenge underlying assumptions, and iterate on models as the network and operating conditions evolve.

Nice to have

Experience contributing to production digital twins or simulation platforms in logistics, e-commerce, or network-heavy domains.

Background in pricing, margin analysis, or commercial optimisation where models directly affect P&L decisions.

Familiarity with logistics operations, transport cost structures, or supply chain constraints that shape network behaviour.

Practical notes

This is a full-time position based in London with hybrid working expectations.

The role sits within a data organisation of approximately 30 engineers, analysts, and data scientists, embedded within the finance squad to ensure tight alignment on strategic objectives.

You will work alongside more senior data scientists who own the broader direction of the digital twin, while a dedicated finance analyst partners with you directly on reporting and visibility layers.

The successful candidate will help shape the roadmap for the digital twin and be expected to ship components that remain reliable and extensible as Relay continues to scale.