Senior Data Scientist - Middle Mile & Pitstops

As a Senior Data Scientist at Relay, you'll be responsible for optimizing our middle mile operations, which includes modeling and improving vehicle use and handover reliability. You'll partner with various teams to enhance driver acquisition and pricing, and work on optimizing pitstop expansion aligned with our growth strategy. Your role will also involve collaborating with colleagues to automate model pipelines and providing insights for strategic decision-making across operational and commercial areas.

To qualify for the Senior Data Scientist position at Relay, candidates should have at least 6 years of experience in data science, demonstrating a successful track record in delivering production-ready models. Strong proficiency in Python and SQL is essential, alongside a solid understanding of statistics and probability applied in operational contexts. Excellent communication skills for explaining technical results to non-technical audiences and the ability to navigate complex data are also crucial.

The Senior Data Scientist role at Relay involves utilizing smart tooling and automation in conjunction with AI to create actionable insights from complex data sets. The position also requires deep familiarity with Python, SQL, and various statistical tools, as well as experience with geospatial data and forecasting models. Collaboration with teams across functions will often lead to using Microsoft Excel or Google Sheets for financial modeling.

At Relay, the work culture is one of learning, ownership, and teamwork. As a Senior Data Scientist, you'll be encouraged to take initiatives from day one, engage in team socials, and participate in exciting events that foster camaraderie. The company values flexibility and offers hybrid working options, ensuring you can maintain a healthy work-life balance while contributing to meaningful projects.

Relay offers a competitive benefits package for the Senior Data Scientist role, including 25 days of annual leave, an equity package, and comprehensive health insurance through Bupa Global. Additional perks include a contributory pension scheme, hybrid working arrangements, free gym membership at the co-working space in London, and a cycle-to-work scheme. This is in addition to a supportive workplace culture that prioritizes employee growth and development.

When optimizing a logistics network, I begin by analyzing existing data to identify inefficiencies or bottlenecks. By developing predictive models using Python and SQL, I assess various scenarios and their potential impacts. It's also vital to collaborate with cross-functional teams to ensure that optimizations align with broader business goals and operational strategies.

In my previous roles, I frequently used predictive modeling to forecast demand and optimize supply chain operations. I focused on employing regression techniques and machine learning algorithms to analyze historical data, which allowed us to anticipate trends and inform strategic decisions. Communication with stakeholders was key in ensuring these models were actionable and relevant.

Working with messy data is a common challenge in data science. I approach this by first assessing data quality and identifying any gaps or inconsistencies. Techniques such as data cleaning, transformation, and even implementing robust data pipelines help improve data integrity. Ensuring I maintain communication with my team helps align on the right approaches to clean and utilize this data effectively.

In one successful project, I led a team in enhancing the delivery efficiency of a logistics network. By developing a multi-variable model that assessed route performance, vehicle utilization, and customer feedback, we identified key areas for improvement. Implementing data-driven changes led to a 20% reduction in delivery times, significantly enhancing customer satisfaction.

My strategy for communicating complex data results involves simplifying technical jargon and focusing on actionable insights. I often use visualizations to illustrate key points, ensuring that stakeholders can grasp the findings quickly. It's all about framing the data in a way that highlights its relevance to their objectives, making them more likely to understand and use the insights effectively.

Financial modeling has been an integral part of my work, particularly in assessing the viability of new projects. I create detailed financial models to forecast revenues and expenses, which allow for evaluating potential risks and returns. Collaborating with finance teams helps ensure these models are accurate and aligned with overall business strategies.

I've worked with geospatial data in optimizing delivery routes and sites for new distribution points. By analyzing geographic patterns and customer density, I used GIS tools to visualize data, which informed the strategic placement of resources. This analysis played a vital role in cutting costs and improving service levels in logistics operations.

To ensure models are ready for production, I follow a rigorous testing and validation process. This includes setting aside test datasets to evaluate model accuracy and performance. Additionally, I collaborate closely with MLE teams to automate deployment, ensuring that our models can be integrated seamlessly into existing systems with minimal disruption.

In a fast-paced environment, I prioritize tasks based on business impact and urgency. I use project management tools to visualize workflows and deadlines, allowing me to allocate resources effectively. Regular check-ins with team members also help ensure we're aligned on priorities and adapting to changes swiftly when necessary.

I am drawn to Relay because of its commitment to sustainability and innovation in the logistics space. The opportunity to apply data science in a meaningful way that directly impacts the community and environment excites me. Moreover, the collaborative culture and focus on personal growth resonate with my professional values, making Relay an ideal fit for my skills and aspirations.