Scientist II, Computational Biology, AI & Multimodal Target Discovery

Scientist II, Computational Biology, AI & Multimodal Target Discovery
JOB DESCRIPTION
Department Data Science, PIRC
Reports to Director of AI in Drug DevelopmentJob Summary
We are seeking a computational biologist to advance target discovery and translational research through integrated multi-omics, agentic AI, computational structural biology, and predictive modeling of cellular perturbations. This individual will build and apply rigorous computational methods that connect human disease evidence to target hypotheses, predict the effects of genetic and pharmacologic interventions, and guide experimental design. The role partners closely with Immunology, Translational Science and Biologics to deliver reproducible analyses, decision-ready evidence, and reusable scientific framework for PIRC's research pipeline.Key Responsibilities
• Integrate and analyze multi-omics datasets, including bulk and single-cell transcriptomics, spatial omics, proteomics, human genetics, functional genomics, and clinical or translational data, to identify disease-relevant cell states, pathways, biomarkers, and therapeutic targets.
• Develop AI agents that support scientific workflows such as literature and knowledge retrieval, data quality assessment, analysis planning, code and tool execution, hypothesis generation, and result synthesis, with appropriate human review, traceability, access controls, and evaluation.
• Develop transparent target-identification and prioritization frameworks that combine genetic evidence, causal inference, knowledge graph, tractability, safety, and translational relevance; clearly communicate evidence strength, uncertainty, and recommended next steps.
• Build, evaluate, and apply virtual-cell and perturbation-response models to predict the effects of genetic and pharmacologic interventions across cell types, disease contexts, doses, and time points; partner with laboratory scientists on prospective validation and model refinement.
• Apply computational structural biology to assess target biology and tractability, including protein structure prediction and quality assessment, variant and domain interpretation, interaction-site analysis, docking, molecular dynamics, and structure-informed hypothesis generation.
• Create reusable, well-maintainable computational frameworks and scalable workflows for data ingestion, harmonization, analysis, model training, and reporting; establish standards for versioning, provenance, quality control, reproducibility, and secure handling of research data.
• Collaborate with Immunology, Biologics, Translational Science, CMC, Clinical, Medical, and other partners to translate program questions into computational plans and deliver actionable results at target nomination and subsequent decision points.
• Track and critically evaluate emerging methods in multimodal foundation models, agentic AI, structural modeling, and computational biology; adopt new approaches when benchmarking shows a clear scientific or operational advantage.
• Communicate methods, findings, limitations, and recommendations through clear visualizations, technical documentation, presentations, and contributions to publications and conferences.Required Qualifications

• Ph.D. in Computational Biology, Bioinformatics, Computer Science, or a related quantitative discipline, and 0-3 years' industry experience.
• Demonstrated depth in at least two of the following areas, with the ability and interest to work across the others: multi-omics target discovery, AI agent development, computational structural biology, or cellular perturbation modeling.
• Demonstrated experience using cutting edge computational biology and ML/AI to answer drug-discovery or translational research questions in a pharmaceutical setting.
• Hands-on experience with multi-omics analysis and biological interpretation; experience with single-cell transcriptomics is required, and experience integrating multiple modalities is strongly preferred.
• Strong programming skills in Python/Java/C#/R, together with sound practices in software design, testing, version control, documentation, and reproducible analysis. Vibe coding under robust code review and management is encouraged.
• Working knowledge of modern ML/DL/AI methods for biological data, including model evaluation, data leakage prevention, uncertainty, and appropriate benchmark design.
• Ability to turn ambiguous scientific questions into testable computational hypotheses and to explain methods, assumptions, limitations, and decisions to multidisciplinary audiences.
• Strong collaborative skills and an understanding of the target discovery, validation, and translational research process, particularly in oncology, hematology, or immunology.

Preferred Qualifications
• Experience with perturbational datasets and methods such as CRISPR screens, Perturb-seq, transcriptomic drug-response resources, multimodal or foundation models, and counterfactual prediction.
• Experience building LLM-based or agentic systems using tool calling, retrieval-augmented generation, workflow orchestration, structured outputs, evaluation harnesses, and human-in-the-loop controls.
• Experience with protein structure prediction, molecular visualization, sequence-to-structure analysis, docking, molecular dynamics, or protein-protein interaction modeling.
• Familiarity with cloud or high-performance computing, containers, workflow engines, relational or graph databases, and production-oriented model or data pipelines.
• A record of scientific publication, open-source contribution, patents, or delivery of computational methods that influenced experimental or portfolio decisions.