Senior Data Scientist - Middle Mile & Pitstops

Relay
London
1 week ago
Create job alert

Company mission

In the future, almost everything we consume will simply materialise on our doorsteps - what we call "e-commerce" today will simply be "commerce" tomorrow. But if we continue on todays trajectory, the growth of e-commerce risks damaging the environment, alienating our communities, and straining the bottom-line for small businesses.

Relay is an e-commerce-native logistics network. We are built from the ground up for environmental, social, and economic sustainability. By building from the ground up we are able to entirely rethink both the middle and last mile enabling us to reduce the number of miles driven to deliver each parcel, lower carbon emissions, and lower costs, all while channelling funds to community members.

At the same time, were fixing the last broken aspect of e-commerce for consumers: delivery. As shoppers, we should have complete control over when and how we receive our purchases, and we should be able to return unwanted items as easily as we ordered them. Thats why whenever you buy from a merchant powered by Relay, youll be able to reschedule your delivery at any time. And if you dont like what you ordered, at the tap of a button well send someone to pick it up.

To orchestrate this complex ballet, Relay relies on a wide range of technologies, from advanced routing and planning to sophisticated user experiences that guide our team members on the ground.
About the role

As a highly operational business, we rely on data science to power nearly every part of our network - from forecasting parcel volumes, to pricing and planning courier capacity, to understanding and improving the economics of our operation.

Were hiring a Senior Data Scientist to help us optimise our middle mile operation and model the growth, performance, and economics of our pitstop network. This role spans across domains, touching forecasting, operations, and commercial planning, and is ideal for someone who thrives on applying models in ambiguous, real-world environments.

Youll work with squads across routing, sortation, first mile, last mile, marketplace, and commercial functions; youll focus on middle mile optimisation, pitstop expansion, and understanding the long-term financial value of our physical network. Youll also bring together data from across the business, often fragmented or messy, and use smart tooling, automation, and AI to transform it into usable insight.

Youll need to be hands-on and pragmatic; its a high-impact role with strong exposure to leadership and decision-making across the business.
What youll do

  • Model and improve the cost, quality, and efficiency of middle mile operations, including vehicle use, timings, and handover reliability
  • Partner with marketplace and ops teams to optimise driver acquisition, targeting, and pricing for the middle mile
  • Optimise pitstop expansion in line with volume growth, capacity, and service levels
  • Model pitstop-level LTV and unit economics to support capital investment and performance tracking
  • Collaborate with other data scientists to support geo-sequencing, zone design, and integration with routing models
  • In partnership with MLE and Staff Data Scientists, orchestrate and automate model pipelines in production
  • Act as a thought partner for operations, commercial, and finance leads - bringing a scientific lens to planning and network growth

What were looking for

  • 6+ years of experience in data science, with a strong record of delivering models into production
  • Deep experience with Python and SQL
  • Strong foundations in statistics and probability, with experience applying them in operational and/or financial contexts
  • Comfort working in ambiguity and navigating messy or incomplete data
  • Effective communication skills - you can explain technical results clearly to non-technical audiences
  • Comfort working across functions and disciplines to drive impact

Nice to haves

  • Experience working in logistics, marketplaces, or similarly complex operational businesses
  • Exposure to business planning, pricing, or commercial decision-making; experience with forecasting, scenario, and financial modelling (including partnering with Finance and Commercial teams and their models (in Excel, Google Sheets))
  • Familiarity with geospatial data
  • Experience in fast-scaling startups or operational teams

What we offer

  • 25 days annual leave per year (plus bank holidays).
  • Equity package.
  • Bupa Global: Business Premier Health Plan - Comprehensive global health insurance with direct access to specialists, dental care, mental health support and more.
  • Contributory pension scheme.
  • Hybrid working
  • Free membership of the gym in our co-working space in London.
  • Cycle-to-work scheme
  • A culture of learning and growth, where youre encouraged to take ownership from day one.
  • Plenty of team socials and events - from pottery painting to life-size Monopoly and escape rooms

J-18808-Ljbffr

Related Jobs

View all jobs

Senior Data Scientist

Senior Data Scientist

Senior Data Scientist (Document Search)

Senior Data Scientist

Senior Data Scientist

Senior Data Scientist

Get the latest insights and jobs direct. Sign up for our newsletter.

By subscribing you agree to our privacy policy and terms of service.

Industry Insights

Discover insightful articles, industry insights, expert tips, and curated resources.

Portfolio Projects That Get You Hired for Machine Learning Jobs (With Real GitHub Examples)

In today’s data-driven landscape, the field of machine learning (ML) is one of the most sought-after career paths. From startups to multinational enterprises, organisations are on the lookout for professionals who can develop and deploy ML models that drive impactful decisions. Whether you’re an aspiring data scientist, a seasoned researcher, or a machine learning engineer, one element can truly make your CV shine: a compelling portfolio. While your CV and cover letter detail your educational background and professional experiences, a portfolio reveals your practical know-how. The code you share, the projects you build, and your problem-solving process all help prospective employers ascertain if you’re the right fit for their team. But what kinds of portfolio projects stand out, and how can you showcase them effectively? This article provides the answers. We’ll look at: Why a machine learning portfolio is critical for impressing recruiters. How to select appropriate ML projects for your target roles. Inspirational GitHub examples that exemplify strong project structure and presentation. Tangible project ideas you can start immediately, from predictive modelling to computer vision. Best practices for showcasing your work on GitHub, personal websites, and beyond. Finally, we’ll share how you can leverage these projects to unlock opportunities—plus a handy link to upload your CV on Machine Learning Jobs when you’re ready to apply. Get ready to build a portfolio that underscores your skill set and positions you for the ML role you’ve been dreaming of!

Machine Learning Job Interview Warm‑Up: 30 Real Coding & System‑Design Questions

Machine learning is fuelling innovation across every industry, from healthcare to retail to financial services. As organisations look to harness large datasets and predictive algorithms to gain competitive advantages, the demand for skilled ML professionals continues to soar. Whether you’re aiming for a machine learning engineer role or a research scientist position, strong interview performance can open doors to dynamic projects and fulfilling careers. However, machine learning interviews differ from standard software engineering ones. Beyond coding proficiency, you’ll be tested on algorithms, mathematics, data manipulation, and applied problem-solving skills. Employers also expect you to discuss how to deploy models in production and maintain them effectively—touching on MLOps or advanced system design for scaling model inferences. In this guide, we’ve compiled 30 real coding & system‑design questions you might face in a machine learning job interview. From linear regression to distributed training strategies, these questions aim to test your depth of knowledge and practical know‑how. And if you’re ready to find your next ML opportunity in the UK, head to www.machinelearningjobs.co.uk—a prime location for the latest machine learning vacancies. Let’s dive in and gear up for success in your forthcoming interviews.

Negotiating Your Machine Learning Job Offer: Equity, Bonuses & Perks Explained

How to Secure a Compensation Package That Matches Your Technical Mastery and Strategic Influence in the UK’s ML Landscape Machine learning (ML) has rapidly shifted from an emerging discipline to a mission-critical function in modern enterprises. From optimising e-commerce recommendations to powering autonomous vehicles and driving innovation in healthcare, ML experts hold the keys to transformative outcomes. As a mid‑senior professional in this field, you’re not only crafting sophisticated algorithms; you’re often guiding strategic decisions about data pipelines, model deployment, and product direction. With such a powerful impact on business results, companies across the UK are going beyond standard salary structures to attract top ML talent. Negotiating a compensation package that truly reflects your value means looking beyond the numbers on your monthly payslip. In addition to a competitive base salary, you could be securing equity, performance-based bonuses, and perks that support your ongoing research, development, and growth. However, many mid‑senior ML professionals leave these additional benefits on the table—either because they’re unsure how to negotiate them or they simply underestimate their long-term worth. This guide explores every critical aspect of negotiating a machine learning job offer. Whether you’re joining an AI-focused start-up or a major tech player expanding its ML capabilities, understanding equity structures, bonus schemes, and strategic perks will help you lock in a package that matches your technical expertise and strategic influence. Let’s dive in.