Head / Principal Data Scientist, Analytics & Machine Learning in England

Summary
Responsibility
: Plan and execute Valink\’s data strategy to build predictive models for drug discovery and drug positioning
Salary
: Competitive salary at Director or Principal level, depending on experience
Starting date
: Jan-Feb ****
Location
: White City, London, UK
About Us
Valink Therapeutics is a spinout from the University of Oxford with a mission to revolutionise the field of bispecific antibody-drug conjugate (bsADC) discovery.
We have developed the most advanced drug discovery platform for complex modalities, capable of generating and screening over 1,000 drug candidates per week to uncover novel, first-in-class bsADCs that others cannot.
Our approach generates a rich dataset capturing cytotoxicity, target expression, payload sensitivity, and related correlations, offering high potential for integration into predictive models that accelerate and enhance future drug discovery efforts.
Role Summary
The Principal Data Scientist will help shape and execute our strategy for integrating proprietary screening data with internal & external biological datasets to identify and prioritise drug candidates for testing.
This role is hands-on: on top of shaping Valink\’s strategic input on data analysis approaches and AI-driven predictive modelling, the profile will also build prototype pipelines, curate relevant datasets, and validate methodologies that will inform the longer-term development of our in-house AI platform.
Key Responsibilities
Driving Data Strategy
Determine appropriate models and computational frameworks for predictive drug-target sensitivity analysis.
Inform on infrastructure, data architecture, and workflow considerations for scalable AI adoption.
Data Curation, Integration and Analytics
Identify, source, and curate publicly available datasets (cell lines, patient data, target expression, protein/compound databases).
Harmonise and integrate these external resources with Valink\’s internal phenotypic and screening data.
Ensure data quality, interoperability, and relevance for downstream predictive modelling.
Perform EDA to uncover patterns, trends and outliers in our data to augment downstream design and modelling processes.
Model Prototyping & Development
Build and test machine learning pipelines to predict correlations between cytotoxicity, target expression, and payload sensitivity.
Explore applications of AI-driven drug positioning approaches to support candidate selection.
Benchmark different models and methods and evaluate trade-offs to derive the best model.
Turn model outputs into clear insights and visualisations that biologists can act on, helping teams move from hit discovery to candidate optimisation.
Collaboration & Knowledge