Data Scientist

You'll join an established and highly experienced team of data scientists and MLOps specialists, and work closely to develop and maintain our suite of models, including:

• Valuation models: Predicting vehicle values across wholesale and retail markets.
• Retail pricing models: Optimising vehicle prices in our retail businesses to maximise profitability.

This is a hybrid role spanning both Data Science and Machine Learning engineering. You'll be involved in the full ML lifecycle - collaborating with stakeholders, developing models that are relevant and impactful and then deploying, maintaining, and improving them in production.

The ideal candidate will have hands-on experience as a data scientist, with strong technical foundations and a curious, humble, and impact-driven mindset. Main Responsibilities:

• Contribute to the evolution of our pricing and valuation strategy by collaborating with stakeholders and peers across the Data Science and MLOps teams.
• Develop innovative approaches to improve the capability and accuracy of our valuation and pricing models.
• Deploy solutions that are robust, scalable, monitored, and aligned with engineering best practices.
• Partner with Analytics, BI, and Data Engineering teams to integrate model outputs into decision-making tools.
• Be an active contributor to the team - sharing knowledge and best practices, experimenting with new techniques and questioning / challenging constructively
• Contributing ideas to the product roadmap that align with business goals and stakeholder needs.

Do you have experience in SQL?, Essential

• Solid experience with Python and SQL
• Experience designing and validating regression models on structured data using appropriate techniques (e.g. linear models, random forests, XGBoost, LightGBM)
• Experience deploying models into production following MLOps practices (e.g. version control, CI/CD)
• Strong attention to detail and commitment to delivering high-quality, well-tested code.
• Familiarity with cloud platforms such as Databricks, Azure, AWS or GCP.
• Ability to contribute to solution design by understanding business context and asking clarifying questions.
• Strong communication skills and ability to explain technical concepts to non-technical audiences
• A self-starter mindset with a willingness to learn, ask questions and collaborate

Preferred (but not essential)

• Experience in pricing, valuation, or retail optimisation domains
• Experience creating and running experiments
• Familiarity with linear optimisation techniques
• Familiarity with collaborative development practices (e.g. peer reviews, boards, sprint planning).

We've put everything in place to ensure you thrive. Competitive salaries, strong core values and an excellent benefits package that includes:

• Company Pension Scheme
• An Employee Assistance Programme
• Enhanced maternity, paternity, and adoption leave
• Access to trained Mental Health First Aiders

The Constellation Automotive Group is the largest vertically integrated digital car marketplace in Europe, combining the leading digital brands across the segments of Consumer to Business (C2B), Business to Business (B2B) and Business to Consumer (B2C). Businesses in our group include house-hold names such as cinch, We Buy Any Car, and Marshall Motor Group as well as industry leading British Car Auctions (BCA). Our Data Science function - now in its tenth year - sits across the Constellation Automotive Group and is at the forefront of providing data products and services that help underpin the success of these companies. Our cross-group remit is broad with our models ranging from valuation and commercial pricing to marketing and personalisation, through to operational optimisation including stock selection, delivery route optimisation and vehicle refurbishment. We're committed to building an inclusive workplace where everyone feels valued and supported. We welcome applications from all qualified candidates of all genders, including but not limited to people of colour, the LGBTQ+ community, people with disabilities, neurodivergent individuals, parents, carers, and those from lower socio-economic backgrounds. Main Purpose of Role: Vehicles are expensive assets and valuing them accurately is incredibly challenging. No two cars are the same - with thousands of models and specifications, plus individual differences in age, condition, and mileage, every vehicle presents a unique valuation problem. We love a challenge though. Our valuation and pricing models underpin key decisions in our own business and across the wider automotive ecosystem - from enabling confident wholesale transactions and accurate part-exchange offers, to helping businesses optimise pricing and assess the most profitable route for each vehicle.