Data Scientist

The RoleWe are looking for a Data Scientist who wants to work on some of the hardest and most interesting problems in healthcare AI.You will work with large-scale, longitudinal EHR data to develop models across areas such as:Clinical prediction and risk stratificationPatient journey and disease progression modellingTreatment and medication intelligenceClinical outcome predictionDoctor and patient behaviour modellingHealthcare recommendation systemsMedical NLP and clinical language understandingData quality and clinical entity modellingFeatures and datasets for foundation and generative AI modelsYou will work closely with ML engineers, clinicians, product managers and data engineers to take ideas from research to production.What You Will Do1. Build healthcare modelsDesign, train and evaluate statistical and machine learning models using real-world clinical data.2. Work with longitudinal EHR dataUnderstand how diagnoses, symptoms, medications, investigations and clinical encounters evolve over time and translate these patterns into meaningful features and models.3. Solve ambiguous problemsTurn open-ended healthcare questions into measurable modelling problems. Decide what data is required, how the target should be defined and how success should be evaluated.4. Develop clinical intelligenceIdentify patterns in patient journeys that can help predict risks, outcomes, adherence, treatment response and disease progression.5. Work with unstructured clinical dataBuild pipelines and models using clinical notes, prescriptions, medical terminology and other unstructured healthcare information.6. Experiment rapidlyDevelop hypotheses, build prototypes, run experiments and iterate quickly. We value strong problem-solving and scientific thinking over simply applying standard modelling techniques.7. Build production-ready solutionsWork with engineering teams to take models from notebooks into scalable production systems and continuously monitor their performance.8. Establish rigorous evaluationHealthcare models need more than good offline metrics. You will design evaluation frameworks that consider clinical relevance, bias, calibration, robustness and real-world performance.What We're Looking For2–6 years of experience in Data Science, Machine Learning, Applied Statistics or a related fieldStrong understanding of machine learning and statistical modellingStrong Python and SQL skillsExperience with libraries such as Pandas, NumPy, Scikit-learn, PyTorch or equivalentStrong understanding of model evaluation, experimentation and statistical reasoningExperience working with large and messy real-world datasetsAbility to translate business or domain problems into well-defined modelling problemsStrong analytical and problem-solving skillsAbility to communicate technical findings clearly to non-technical stakeholdersStrong PlusExperience in one or more of the following will be highly valuable:Healthcare, EHR or claims dataClinical NLP / medical language modelsTime-series or longitudinal modellingSurvival analysisCausal inferenceRecommendation systemsRisk predictionGenerative AI / LLMsMedical terminology and ontology systemsLarge-scale data processing using Spark or similar technologiesBuilding and deploying ML models in productionWhat Makes This Role DifferentReal clinical data.You will work with longitudinal healthcare data generated through actual clinical workflows, not synthetic datasets or toy problems.Hard problems.Healthcare data is noisy, incomplete, heterogeneous and highly contextual. Building useful models requires going beyond standard ML approaches.Large opportunity.We are building a foundation for healthcare AI across millions of patient journeys and millions of clinical interactions.Research meets product.You will have the freedom to explore new modelling approaches while also seeing your work deployed into products used by doctors.Clinical impact.The ultimate measure of our models is not just AUC or F1. It is whether they can improve clinical decision-making and healthcare outcomes.Our Ideal CandidateYou are someone who gets excited when you see a messy dataset and immediately start asking:What is the underlying pattern?What could we predict from this?Is there a better way to represent this clinical journey?Can we build a model that actually generalises to the real world?You enjoy going deep into a problem, challenging assumptions and building things that have never existed before.If you want to work at the intersection of EHR data, machine learning, clinical intelligence and generative AI, this is the opportunity. (ref:hirist.tech)