Head of Clinical Data Science and AI Research

We are looking for an accomplished healthcare data scientist to lead the scientific and technical work across Barnabus Care. You will bring a substantial publication record, strong practical engineering skills, and experience taking research into clinical evaluation or operational use.

You will own the data science roadmap, research methods, evaluation quality, and development of the research team. A substantial part of your work will remain technical, including reviewing code, investigating data, designing studies, and resolving difficult methodological problems.

You will work directly with the CEO and CTO, Chief Medical Officer, Chief Clinical Information Officer, engineering leadership, product teams, and Publications Office. Clinical leaders own clinical approval. Engineering leadership owns production architecture and operations. You will be accountable for the scientific and technical quality of the data science work that supports both.

What you will be responsible for
Scientific direction and team leadership
Set research priorities with clinical, product, and engineering leaders. Define the intended use, evidence requirements, milestones, and accountable owners for each priority.
Hire, mentor, and review data scientists and researchers. Give specific feedback on code, analyses, protocols, and written work, and address weak work early.
Maintain clear technical documentation, reproducible analyses, and shared ownership so colleagues can understand, challenge, and continue the work.

Clinical data and technical implementation
Lead work with longitudinal health records, clinical notes, laboratory results, observations, medications, procedures, and outcomes. Resolve inconsistencies in timing, units, terminology, and source records.
Build and review Python and SQL pipelines for clinical data preparation, statistical analysis, machine learning, and natural language processing. Choose rules, statistical methods, retrieval, or language models according to the clinical task.
Develop reusable research pipelines, analytical methods, and evaluation tools. Work with engineering on documented interfaces, versioned artifacts, testing, monitoring, and reliable integration.
Evaluate language model outputs for factual accuracy, omissions, temporal errors, unsupported conclusions, and source attribution. Define how systems handle conflicting information, uncertainty, and cases that require clinician review.

Clinical validation and research quality
Design studies with explicit cohorts, eligibility criteria, reference standards, endpoints, sample size justification, and analysis plans. Establish clinician review and adjudication procedures before evaluation begins.
Prevent patient and temporal leakage. Assess selection bias, missing data, label quality, subgroup performance, and generalizability across clinical settings. Use independent and external evaluation where the intended use requires it.
Define acceptance criteria with clinical owners. Evaluate clinical errors, uncertainty, and workflow effects alongside task-appropriate measures such as sensitivity, precision, calibration, and agreement.
Evaluate analytical methods and AI systems against their approved specifications, including missing information, contradictory inputs, boundary cases, and version changes. Preserve the link between each finding and its source evidence.
Plan retrospective studies, prospective evaluations, and monitored clinical pilots with hospital partners. Distinguish results from synthetic scenarios, retrospective datasets, and live clinical use when reporting what the system can do.

Publications and scientific partnerships
Develop original research questions that address real product and clinical problems. Work with the Publications Office on protocols, manuscripts, conference submissions, and responses to reviewers.
Make reported findings reproducible through documented datasets, analysis code, experiment records, and explicit limitations. Ensure authorship reflects actual contributions.
Work directly with clinical investigators and research ethics teams on study approvals, permitted data use, and research execution. Communicate progress and unresolved issues clearly.

Delivery and accountability
Turn research findings into implementation specifications and measurable delivery plans with engineering and product teams. Track dependencies and raise blockers early.
Review performance changes after model, rule, prompt, or data updates. Investigate failures and agree corrective actions before extending clinical use.
Present recommendations to leadership with the supporting evidence, limitations, and decisions required. Challenge unsupported conclusions and correct mistakes promptly.