Director, Data Scientist - Clinical AI

Locations: Cambridge, UK or Barcelona, Spain

Salary & Benefits: Competitive

About AstraZeneca and AISI

At AstraZeneca, technology and science meet to change what is possible for patients. We are building a connected, end-to-end Enterprise AI engine — uniting data foundations, AI technology, process reinvention, and business-facing AI to accelerate results across the whole value chain. Success depends on being exceptional connectors: you will actively leverage existing capabilities, celebrate and promote reuse, export breakthrough ideas across geographies and functions, and obsess over scaling impact rather than building in isolation.

AI Science & Innovation (AISI) sits at the centre of AstraZeneca's R&D AI transformation. Our remit is to build, buy, and deliver the AI models and agents that change pipeline outcomes across discovery, translational science, biomarkers, and clinical development.

Within AISI, the Clinical AI team are building world-class AI capability to accelerate the design, conduct, and analysis of clinical trials across our BioPharmaceuticals pipelines — spanning both early and late phase programmes. We partner closely with clinical development, regulatory, and biometrics teams to bring better treatments to patients faster, while adhering to the highest evidentiary standards.

The Opportunity

Bringing new treatments to patients demands scientific excellence at every stage of development. In the Clinical AI team, we focus on one of the most data-rich and decision-intensive parts of that journey: clinical development. Trial design, patient selection, dose optimisation, biomarker strategy, and safety evaluation each represent genuine opportunities where AI and machine learning can add rigour, speed, and precision — not as a replacement for clinical and statistical expertise, but as a powerful complement to it. We hold ourselves to measurable standards of improvement, and we build methods that can be evaluated, reproduced, and trusted in regulatory settings.

You will work across the enterprise to define and deliver on AstraZeneca's most pressing clinical development questions — leading cross-functional teams spanning the key BioPharmaceuticals disease areas of cardiovascular, renal, metabolic disease, respiratory, immunology and cell-therapy. You and the team will develop reusable methods and enterprise-scale approaches that measurably advance the late-stage drug pipeline. This is a high-visibility opportunity to shape how AstraZeneca does AI forBioPharmaceuticals clinical development — from methodology standards to external scientific influence.

AI for clinical development is a field in motion. Foundation models, agentic systems, and causal AI are advancing rapidly, and the regulatory and methodological frameworks around them are evolving in parallel. As a Director, Data Scientist, you will define and drive the AI methodology agenda for one or more programmes within Clinical AI, leading by scientific influence and matrix coordination rather than through a formal hierarchy. You will be the scientific authority that study teams, biometrics, and regulatory colleagues turn to — and AstraZeneca's voice externally at the critical moment when the rules of the road for AI in clinical trials are being written.

Key Responsibilities

• Define and drive the AI methodology roadmap for assigned Clinical AI programmes, spanning early and late phase clinical development, and aligning AI/ML priorities with clinical and business objectives.

• Lead, by matrix influence and scientific authority, the delivery of the most complex and high-stakes AI projects — from problem definition and methodology selection through validation, regulatory alignment, and scaled adoption across the enterprise.

• Develop and govern reusable, enterprise-grade AI methods and evaluation frameworks for clinical trial settings, including innovative trial design support, dose optimisation, biomarker discovery, digital twins, predictive and prognostic modelling, and safety and efficacy signal detection.

• Champion data-centric AI practices at programme level: govern the acquisition, curation, and quality control of datasets for model training, post-training, benchmarking, and evaluation across clinical and regulatory settings.

• Partner with Clinical Development, Biometrics, Regulatory, and Study Teams to embed AI strategy and validated solutions into study design and decision-making at programme level.

• Shape the AI evidence component for regulatory submission packages; act as the scientific and methodological voice in regulatory engagements involving AI/ML methods or innovative trial designs (FDA, EMA, MHRA).

• Evaluate and champion cutting-edge AI methodologies — including foundation models, agentic AI systems, generative patient models, multimodal learning, Bayesian inference, causal inference, and model calibration and domain adaptation — proposing fit-for-purpose approaches with robust evaluation criteria and risk assessment.

• Establish and maintain external collaborations with academic institutions, technology partners, and industry consortia to access novel capabilities and advance the scientific agenda.

• Represent AstraZeneca at scientific conferences, standards bodies, and peer-reviewed venues; contribute first- or last-author publications in leading ML and clinical AI journals.

• Serve as a technical mentor and thought partner for Associate Directors and Senior Data Scientists within the Clinical AI team; promote scientific rigour, reuse, and a culture of learning in public.

• Contribute to the broader AISI AI for Clinical Development strategy, including cross-functional ways of working, tooling governance, and methodology standards.

Essential Requirements

• PhD in Computer Science, Machine Learning, Statistics, Mathematics, Biomedical Informatics, Computational Biology, or a closely related quantitative discipline — with a strong, hands-on computational track record.

• 4–8 years of post-PhD experience in AI and machine learning method development, with demonstrated and sustained impact in clinical, biomedical, or drug development settings (e.g. models delivered, first-author publications, patents, SaMD filings, open-source projects).

• Deep experience, knowledge, and understanding of one or more fields of biology, with hands-on experience working with biological data such as molecular (DNA, RNA, protein), imaging (radiology, microscopy), or clinical text (EHR, clinical notes).

• Deep expertise in modern AI methodologies, including one or more of: foundation model training and fine-tuning; Bayesian inference; temporal and longitudinal modelling; multimodal integration; model calibration and domain adaptation; data-centric AI; model interpretability; model post-training and alignment.

• Exceptional software engineering skills: Python, deep learning frameworks (e.g. PyTorch), frontier coding agent frameworks, modern LLM tooling, and cloud platforms (e.g. AWS, Azure, GCP).

• Demonstrated experience translating AI methods into applications that inform clinical and/or biomedical decisions, including prospective evaluation or contribution to submission-relevant evidence.

• Track record of driving scientific influence across cross-functional communities — ML, clinical, biostatistics, regulatory — without relying on formal authority.

• Peer-reviewed publications in clinical AI, computational drug development, or leading ML venues (e.g. NeurIPS, ICML, ICLR, Nature Medicine, Lancet Digital Health).

• Excellent written and verbal communication skills, with demonstrated ability to translate technical findings for clinical, regulatory, and executive audiences.

Desirable Skills and Experience

• Direct industry experience in early or late phase BioPharmaceuticals clinical development, including clinical trial design, biomarker discovery, companion diagnostics, dosing, safety, endpoints, or regulatory processes.

• Direct experience contributing to FDA, EMA, PMDA, or MHRA submissions involving AI/ML methods or innovative trial designs.

• Prior regulatory engagement on AI methodology, complex innovative trial design, or AI/ML qualification opinions.

• Experience with MLOps and LLMOps at scale, including CI/CD pipelines and enterprise deployment.

• Open-source contributions, workshop organisation, or standards-body participation.

• Knowledge of computing hardware and its impact on model training and inference at scale.

• Experience in a complex global organisation spanning multiple sites and therapy areas.

• Strong proficiency in augmenting — but not supplanting — daily knowledge work with agentic AI tools.

• Proactively up-to-date with the latest AI research; tries out new tools and methods of interest without waiting to be directed.

• Team-oriented mindset: does what is best for the team and the programme, not just the individual deliverable.

• Ability to deliver high-quality, high-impact contributions independently and at pace.

• Comfort with ambiguity and an instinct to learn in public, prototype early, and fail forward.

• ...