Overview
Our Data Science team focuses on the development of Machine Learning and Deep Learning solutions to solve business problems and deliver actionable insights. This role will lead the development of proprietary deep learning models to enhance our existing underwriting capabilities, improving predictive power and generating high‑impact customer insights.
Responsibilities
* Lead the development of next‑generation deep learning approaches to advance our current underwriting models and keep our core lending capabilities at the forefront of the industry.
* Unlock the value in non‑traditional datasets by building sophisticated neural networks (e.g., LSTMs, RNNs, or Transformers) and finding novel ways to transform raw, multi‑modal inputs into powerful predictive features.
* Collaborate with business stakeholders to prioritize initiatives and bridge the gap between ambitious R&D and tangible in‑market results, driving ideas from prototypes to production.
* Use a combination of business acumen, coding and statistical skills to navigate large amounts of data and extract actionable solutions.
* Work cross‑functionally on projects that support key business initiatives and drive sustainable growth.
Qualifications
* Strong experience developing and deploying deep learning models, particularly for sequential data (e.g., time series or language models).
* A proven track record leading model development, including setting the technical direction, project management, stakeholder communication, and mentoring junior team members.
* Experience producing and managing reliable, maintainable code in SQL/Python in a team setting, including code reviews and setting software engineering best practices.
* Hands‑on experience with modern Machine/Deep Learning frameworks such as PyTorch, TensorFlow, or Hugging Face Transformers, including training models on GPUs or GPU clusters.
* Experience working with structured and unstructured data, such as text, logs, or time series and tokenisation techniques.
* A strong understanding of probability, statistics, machine learning and familiarity with large data set manipulation.
* A drive for continued learning through an internal and external focus, and an ability to prototype new techniques to assess value.
Where and how you'll work
This is a permanent position based in our Nottingham office. We have a hybrid working model: you’ll be based in the office three days a week (Tues‑Thurs) and can work from home on Monday and Friday. Many of our associates have flexible working arrangements, and we are open to discussion to find an arrangement that works for you.
What’s in it for you
* Reward and recognition for contributions to our transformation roadmap.
* High performers enjoy strong and diverse career progression opportunities, with investment in training through Capital One University and external providers.
* Access to core benefits including pension scheme, bonus, generous holiday entitlement and private medical insurance, plus flexible benefits such as season‑ticket loans, cycle‑to‑work scheme and enhanced parental leave.
* Work in open‑plan spaces with accessible facilities, including a fully‑serviced gym, subsidised restaurant, mindfulness and music rooms at our Nottingham head‑office.
Recruitment & diversity
We pride ourselves on hiring the best people, not the same people. Building diverse and inclusive teams is the right and smart thing to do. We want to work with top talent: whoever you are, whatever you look like, wherever you come from. We focus on what you do, not just what you say, and we make our recruitment process fair and accessible.
We are partners with organisations such as Women in Finance, Race At Work Charters, Stonewall and upReach to help people from every walk of life thrive with us. Internal networks include REACH, OutFront, Mind Your Mind, Women in Tech and EmpowHER, providing support and opportunities for associates.
Capital One is committed to diversity in the workplace. If you require a reasonable adjustment, please contact us; all information will be kept confidential and will only be used for the purpose of applying a reasonable adjustment.
#J-18808-Ljbffr