Project Data Scientist

Location
Work from home in Ireland.
Eligibility to work
Unfortunately we cannot offer Irish work permits or Visa sponsorship.
About this Role
As a Project Data Scientist in Research Pathology at Deciphex, you will prepare, characterise, and validate the datasets and model outputs that drive our research pathology AI across toxicologic pathology and translational research. Working at the interface of histopathology, computational pathology, and data science, you will turn whole-slide images and structured study data into curated, statistically characterised datasets and qualified analytical outputs. This will include control-dataset characterisation, confounder and batch-effect detection, data-quality assessment, and validation of downstream AI models.
You will support the evaluation of pathology AI models and downstream applications such as anomaly detection, lesion identification, segmentation, and whole-slide classification. This will include contributing to practice-relevant benchmarks, characterising model performance and failure modes, identifying confounders and batch effects, and applying clinically and operationally meaningful measures of performance.
You will work alongside pathologists, the AI team, software and product teams, contributing to validation strategy and ensuring that datasets and analytical outputs meet the evidentiary and regulatory expectations of preclinical and translational research.
A major focus of the role will be supporting Deciphex’s delivery within two EU-funded public-private partnerships: BIGPICTURE (IMI) and VICT3R (IHI). You will contribute to shared data models, metadata standards, dataset onboarding, control-data analysis, and interoperability between Deciphex platforms and consortium infrastructure. Working across internal teams and external partner organisations, you will help resolve data and transfer issues and ensure that project outputs meet agreed standards for quality, traceability, and responsible data handling.
Key Responsibilities
Support Deciphex’s technical and scientific delivery across BIGPICTURE and VICT3R, working with internal teams and consortium partners to deliver agreed data-science activities
Curate, characterise, and analyse large preclinical and translational pathology datasets, including whole-slide images, structured study data, annotations, and associated metadata
Contribute to the VICT3R data model for whole-slide imaging, including metadata schemas, annotation standards, controlled terminology, and linkage to SEND domains
Characterise control datasets through descriptive and statistical analysis, including trend assessment and detection of confounders and batch effects relevant to Virtual Control Group development
Support interoperability between Patholytix and BIGPICTURE infrastructure by contributing to data mappings, requirements, specifications, and gateway testing
Develop and maintain reproducible data-processing pipelines, quality-control routines, and validation workflows supporting dataset curation, model evaluation, benchmarking, and project delivery
Support the evaluation of downstream pathology AI models through performance characterisation, failure analysis, and clear evidence packages
Support the onboarding of non-clinical whole-slide images and metadata from consortium contributors, including data preparation, validation, transfer, and troubleshooting
Produce technical analyses, documentation, data deliverables, and progress reporting in accordance with agreed consortium milestones
Work with pathologists, AI, software, product, and project teams to translate scientific and operational requirements into reliable datasets and analytical outputs grounded in appropriate pathology context
Apply appropriate data-quality, traceability, GDPR, and GLP-aligned handling practices across consortium submissions and analytical outputs
Required Skills and Experience
Advanced degree in data science, bioinformatics, computational biology, biomedical science, biostatistics, statistics, computer science, or a related quantitative field, or equivalent applied experience
Proficiency in statistical computing and data manipulation using one or more relevant languages and tools
Experience applying data science or machine learning to biomedical imaging, life-science, or other complex scientific datasets.
Experience analysing and evaluating machine-learning model outputs, including performance characterisation, error analysis, and interpretation of results in a scientific or biomedical context
Experience curating, integrating, quality-controlling, and documenting datasets and associated structured metadata
Strong grounding in applied statistics, including exploratory analysis, experimental design, identification of confounders and batch effects, and selection of meaningful performance measures
Ability to translate scientific and project requirements into well-defined datasets, analytical workflows, quality-control checks, and clear evidence-based outputs.
Experience developing reproducible analytical workflows using version control, testing, environment and configuration management, data provenance, and technical documentation
Excellent written and verbal communication in English, including the ability to produce clear technical documentation and communicate data issues to specialist and non-specialist stakeholders
Demonstrated ability to work effectively across multidisciplinary and multi-organisation teams, managing competing scientific and delivery priorities
Desirable Skills and Experience
Understanding of preclinical toxicology study design, treatment and control groups, endpoints, and histopathology workflows
Familiarity with non-clinical data standards and regulated research environments, including SEND/CDISC terminology, GLP principles, or regulatory-grade whole-slide image management
Familiarity with whole-slide imaging, digital pathology platforms, image formats, and image-data management
Prior experience contributing to EU-funded consortia (e.g. IMI, IHI, Horizon Europe) or other multi-organisation scientific partnerships involving industry, academia, CROs, or public bodies
Understanding of biomedical data governance, including data integrity, tracea

Employment type
Full-Time,Permanent

Work arrangement
Yes