Machine Learning Scientist
On-siteOxford, England, United Kingdom
Job Summary
Design, develop, test, and maintain machine learning and AI software for the CODAS project, contributing to sovereign AI ecosystem components including ML services, data pipelines, and deployment-ready applications. Support the translation of research outputs into robust systems while adapting AI approaches to privacy-sensitive, security-conscious, and regulated contexts. Troubleshoot technical and integration issues during development and deployment, and collaborate with internal colleagues and external partners to understand requirements and deliver priorities. Create publishable material, attend conferences, and communicate technical trade-offs to stakeholders. Work within a multidisciplinary environment to integrate solutions into project infrastructure and improve internal tooling and engineering standards.
Required Qualifications
- A Master's degree or PhD in Computer Science, Machine Learning, Statistics, Mathematics, or a closely related discipline, or equivalent research or industry experience
- Established work experience in relevant field
- A strong and demonstrable track record in applied ML/AI research or development, with recognised expertise in one or more relevant areas such as privacy-preserving ML, generative AI, large language models, or federated learning
- Deep understanding of modern ML methods, their assumptions and limitations, and the ability to reason about appropriate application in novel or constrained settings
- Software engineering foundations in Python, including ML frameworks, version control, CI/CD, and reproducible experimental practices
- Skills handling challenging data that requires understanding, preparation, oraganisation and augmentation before use for the purposes of ML/AI research or development
- Experience working across the full ML lifecycle — from problem framing, experimentation and prototyping through to evaluation, integration and deployment
- Ability to contribute to scientific publications, technical reports, and the dissemination of research findings to a wide audience
- Communication skills, with the ability to explain technical concepts and trade-offs clearly to both technical and non-technical audiences
- Ability to work effectively in technically complex and ambiguous environments involving multiple stakeholders
Desired Qualifications
- Experience with privacy-preserving ML techniques such as differential privacy, federated learning, secure computation, or membership inference resistance
- Experience with generative AI systems, large language models, agentic AI approaches, or evaluation methodologies for frontier AI
- Experience working in a multi-partner programme involving industry, government or academia
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.