Senior Data Scientist

Relay’s network is scaling fast. The decisions that shape that growth-where to expand, how to price, and where to invest-depend on models that simulate how the network behaves and evolves as density, geography, and operating models change.

We already have an MVP tool in place. Now we’re looking for someone to help productionize it and build the next generation of strategic models on top of it.

You’ll be a core contributor to Relay’s Digital Twin-a network simulator that captures our unit economics end-to-end, from first-mile collection through sortation, middle-mile, and last-mile delivery.

The Digital Twin is already used by Finance, but it’s a living system. Every operating model change, new service type, or commercial scenario requires upgrading a component, adding a model, or creating new ways to analyze the business. Your role is to keep it accurate, scalable, and ready to answer the next strategic question.

What You’ll Do

Your role is split roughly equally across two areas:

1. Evolve the Digital Twin

You’ll develop a deep understanding of every cost component, identify where existing models fall short, and improve them systematically.

You’ll:

• Upgrade the first-mile engine as new operating models roll out.
• Model new service types and their impact on sortation costs.
• Add new network flows and operational metrics.
• Strengthen the parts of the simulator that drive the most important business decisions.

1. Build Strategic Models & Forecasts

Alongside the Digital Twin, you’ll build forecasting and decision-support models that help the business plan for the future.

This includes:

• Volume forecasting
• Predictive modelling
• Scenario planning tools
• Financial models
• Simulation
• Machine learning where it meaningfully improves outcomes

You’ll also help determine which models belong inside the Digital Twin and which should exist independently.

Who You’ll Work With

Your primary partners will be our Finance teams, including:

• Strategic Finance
• Commercial Finance
• FP&A
• You’ll translate business questions into modelling problems and build tools that allow Finance to explore pricing, margins, and forecasting scenarios dynamically-rather than relying on manually rebuilt analyses.
• You’ll help shape the roadmap by understanding stakeholder needs, prioritizing opportunities, and shipping impactful solutions.
• You’ll join a Data organization of around 30 engineers, analysts, and data scientists, while being embedded within the Finance squad.
• You’ll work closely with a dedicated Finance Analyst who owns the reporting and visualization layer built on top of your models, alongside senior Data Scientists responsible for the broader direction of the Digital Twin.

• A systems thinker
• You naturally break complex problems into components, understand the assumptions behind them, and know when those assumptions need revisiting.
• A builder
• You don’t wait for detailed requirements. You investigate problems, determine what’s needed, build useful solutions, and iterate quickly.
• Technically strong
• You’re fluent in Python and SQL and are comfortable owning your own data engineering-from extracting and transforming data through to modelling.
• An engineer at heart
• The Digital Twin is a production system built with Python, SQL, APIs, and a frontend.
• You don’t need to be a frontend expert, but you should be comfortable working in a production codebase-writing clean, tested, maintainable code that others can easily build upon.
• Experienced in modelling
• You have experience with financial modelling, forecasting, simulation, or predictive modelling in environments where your work directly influenced strategic or commercial decisions.
• You understand machine learning and know when it’s worth using-and when it isn’t.
• A strong communicator
• Finance depends on your models to make decisions. You can clearly explain how they work, what assumptions they make, and where they’re reliable (or not).
• Pragmatic
• You care more about solving the business problem than using a particular technique.
• Your goal is to build models that are accurate, useful, and fast-not academically elegant.
• An owner
• You take responsibility for outcomes, manage trade-offs effectively, and care whether your models genuinely improve business decisions.
• Experience in logistics or delivery networks is a plus-but not essential.
• What’s most important is the ability to quickly learn a complex operational domain and build models that help the business make smarter decisions.