Senior Scientist, AI for Translational Sciences

About the role

AstraZeneca is looking for a Senior Scientist for its AI for Translational Sciences team in Barcelona, working 3 days onsite and 2 days from home. The role supports strategic initiatives in machine learning and bioinformatics analysis of molecular data, including DNA sequencing, single-cell and bulk transcriptomics, proteomics, spatial and digital pathology. It suits specialists in AI/ML for molecular data analysis who want to contribute to Oncology and Biopharmaceuticals R&D.

Key responsibilities

• Design and apply innovative computational and AI analysis methods to advance the research portfolio
• Build AI modelling solutions and agentic systems to interrogate diverse molecular datasets
• Empower researchers to use AI systems for data-driven decision-making, translational science and biomedical knowledge generation
• Work within a team maintaining the highest quality standards

Required profile

• Master's degree in Bioinformatics, Computational Biology, Machine Learning or a related field with 4+ years of applied experience in ML/AI molecular data analysis, or a Ph.D. in a related field
• Deep understanding of at least one type of molecular profiling data (e.g. RNA or DNA sequencing) and experience analysing human Illumina NGS data
• Understanding of drug discovery biology: target identification, target validation, mechanism of action or translational science
• Experience using AI tools and agentic frameworks such as coding copilots and research assistants
• Experience working in a team environment

Required skills

• AI/ML modelling approaches to molecular (omics, imaging) data
• Statistical and model-evaluation skills: baseline selection, patient-level and cross-study validation, prevention of data leakage, handling of confounding and batch effects, assessment of uncertainty and generalisability
• Hands-on implementation of deep-learning models with PyTorch, transformers and huggingface_hub
• Proven Python programming and software development best practices (unit testing, version control with git)
• Unix/Linux environments and package management (e.g. Conda)
• Cloud computing (preferably AWS) or high-performance computing (HPC)
• Project leadership and strong communication skills

Desirable: experience with patient data from clinical trials and biomarker development, biological foundation models such as BulkFormer, scGPT and DNABERT, explainability of deep learning models on multi-omics data, and experience in pharma or biotech R&D. Applications close on 29 November 2026.