About the team

Predictive AI and Data teamis responsible forproviding AI and Bioinformatics solutions to the scientists across the spectrum of drug development and discovery in AstraZeneca (both pre-clinical and clinical stages). The primary aim is to find ways to accelerate the drug development process byleveragingexisting company data in combination with the mostcutting-edgeAI approaches in day-to-day scientific work across the company.

Introduction to role

The Associate Director will apply cutting edge advanced computational pathology and AI technologies to multimodal digital pathology datasets and help derive biological insights from combination of multimodal model inference from pathology,omics(transcriptomics, cellpaintintand proteomics)and chemical/ADME foundation modelsfordrug safety prediction to support drug portfolio project milestone decision making.

Accountabilities

• Support Computational Pathology foundation model development:datapreparation including stain/scan harmonization & standardization, data augmentation, rigorous QC,WSItiling/patch extraction.

• Build and evaluate multi-resolution architectures (hierarchical, pyramid, or attention-based models) that integrate tile and slide-level context across 5×-40× magnifications for robust morphological feature learning.

• Undertake self-supervised representation learning: Train foundationmodelson multi-scaleWSIs using scalable distributed training and modernpracticestailored to histopathology.

• Implement, whereappropriate, weak/multiple-instance learning (MIL): Implement andoptimizeMILand weakly supervisedstrategies for downstream tasks where only slide- or case-level labels are available.

• Operatemulti-GPU/cloudenviromentsensuring reproducibility at scale:Distributed learning across GPUs,containers, orchestration, feature stores, versioning, experiment tracking, and checkpointing.

• Evaluation benchmarking for representation quality (linear probe, few-shot, retrieval), cross-site generalization, stain/scan robustness, and uncertainty calibration with statistically sound comparisons.

• Integrate multimodalrepresentations: Investigate approaches to integrate/fuse pathology, Cell Painting, cell-assayomicsand chemical/ADME featurese.g.via joint embeddings, late fusion, and cross-modal attention.

• Support development towards unified drug safety risk scores usingensembling/meta-learning, modality ablations, OOD detection, and prospective validation on preclinical toxicity datasets.

• Collaborate closely with pathologists, computational safety AIscientistsand other safetySMEs;drive interpretability reviews, clear communication to leadership, and milestone planning.

• Scientific communication: Drive high-quality and open science &logisticscommunication across project team to ensure whole team is aligned throughout project

Essential Skills/Experience

• PhD in Computer Vision & AI including computational pathology

• Stronghands-onML/DL experience (preferably usingPyTorch).

• Experience with self-supervised learning/MIL model training approaches

• Advanced Python programming skills.

• Experience with pathology Whole Slide Image (WSI) datasets and preparation & use for model training

• Proven delivery of computational pathology model impact.

• Excellent communication and stakeholder engagement skills.

• Solves complex problems using scientific judgement.

• Builds strong cross-functional relationships and networks.

Desirable Skills/Experience

• Experience with Omics, CellPaintingor otherbiomedicalimaging data.

• Background knowledge or experience working with pathology or similar medical imaging datasets.

• Experience with cloud computingandworkflow automation.