Applied Data Scientist

At intoDNA, we develop STRIDE, the first platform technology for direct detection of DNA damage. It has been successfully used by leading biotech and pharmaceutical companies and renowned academic groups to accelerate their drug development efforts. We're looking for an Applied Data Scientist to join our Data Science & Analysis team. Clients increasingly bring us questions that require non-standard analysis — new assay readouts, unusual experimental designs, tissue samples, multi-parameter endpoints, and clinical studies where STRIDE results must be reconciled with patient outcome data. This role owns those analyses, and turns the recurring ones into validated methods the wider team can reuse. This is a delivery role throughout. You'll work on live commercial and clinical projects — the complex end of our portfolio — as well as internal R&D projects, working directly with Project Owners and, where useful, their clients. New analytical methods are developed here because a project needs them, not in separate research time: when no existing approach fits the question, building one is how you deliver it. This role suits someone strong at quantitative analysis and applied statistics, who has developed analytical methods rather than only applied them, and who wants that work anchored to a real question and a real deliverable. You will also be the person analysts come to when a method or statistics question goes beyond the standard playbook. Responsibilities Own the analytically demanding end of our project portfolio: non-standard endpoints, novel assay readouts, unusual experimental designs, multi-parameter analyses, and tissue-based samples Own the analytical workstream of clinical projects — integrating STRIDE readouts with clinical and patient outcome data to support the science team's work on STRIDE's predictive value Advise Project Owners at scoping on whether a question is answerable with the data proposed, what endpoints and sample sizes are needed, and what the analysis can and cannot conclude Apply statistically sound analysis to hierarchically structured data (cells within images, images within samples, samples within subjects), and produce publication-quality figures and client-ready reporting Develop and validate new analysis methods where a project needs one and none exists — derived metrics, classification schemes, scoring approaches, validation strategies — and document them before the project closes, so another analyst can apply them unaided and their limitations are clear Collaborate with our Data Scientists to turn valuable methods into permanent features of the analysis platform Act as the analytical escalation point for analysts — reviewing approaches, answering method and statistics questions, and mentoring on analytical judgment Requirements - Must have Master's or PhD in a quantitative or life science field with a strong quantitative component (bioinformatics, computational biology, biophysics, biomedical engineering, physics, applied mathematics, or related) Around 3+ years of hands-on experience in quantitative analysis of microscopy or other biomedical imaging data, in industry or a research/core-facility setting Strong Python for the full analysis workflow (pandas, NumPy, SciPy, scikit-image, and matplotlib/seaborn or equivalent), and Git for version control Advanced applied statistics, beyond t-tests and ANOVA: experimental design, power and sample size reasoning, hierarchical or mixed-effects models, multiple comparison correction, effect size estimation. Able to explain why a method is appropriate, not only how to run it Demonstrated experience developing and validating a new analysis method, metric, or scoring approach from scratch — not only applying existing pipelines Working understanding of fluorescence and confocal microscopy and 3D image stacks, and of image segmentation and feature extraction and their common failure modes Ability to work directly with non-computational scientists: translating a biological question into an analysis plan, and communicating results and their limitations honestly Ability to manage your own priorities across several concurrent projects, and to say when something is not achievable in the time available Professional working proficiency in English (written and verbal) Requirements - Nice to have Machine learning applied to image or single-cell data, supervised classification in particular (scikit-learn; PyTorch a bonus) Experience with tissue image analysis or digital pathology (FFPE, IHC/IF, whole-slide imaging) Clinical or translational data analysis: survival analysis, ROC/AUC, biomarker cut-off determination, or predictive biomarker validation Familiarity with DNA damage response, DNA repair, or cell cycle biology Experience mentoring or reviewing the work of more junior analysts Experience in a GxP or otherwise regulated environment What you'll work with Data: Single-cell quantification from fluorescence microscopy — signal intensity, morphological measurements, and derived biomarker scores across thousands of cells per project. Increasingly, clinical samples with associated patient metadata and outcome data Platform: intoDNA's internal microscopy image analysis system, developed in-house and actively evolving Tools: Python (pandas, NumPy, SciPy, scikit-image, scikit-learn), Jupyter, Git, AWS Collaboration: Project Owners and the science team on live projects; Data Scientists, with whom you'll turn new methods into permanent platform features; analysts you'll support and mentor Direction: Work on clinical studies that support intoDNA's development of STRIDE as a predictive biomarker and companion diagnostic What we offer Opportunity to make a real impact in an early-stage, highly innovative company Collaborative work in a stimulating and friendly environment Participation in Employee Stock Options Plan Benefits package, including private medical healthcare and an on-site gym