Senior Data Scientist – Perturbation Biology & AI

Career Category Research Job Description What you will do Let’s do this. Let’s change the world. We are seeking a highly qualified and motivated Senior Data Scientist with a strong background in computational biology to join the Bioinformatics Technologies team within Amgen’s Automation, Research Data Systems, Informatics, and AI (ARIA) organization. ARIA is a multidisciplinary group embedded within Amgen’s discovery engine, leveraging advancements in digital technologies for disease modeling and digital modality engineering to accelerate the pipeline from target inception through drug development. Within ARIA, Bioinformatics Technologies serves as an innovation hub for developing, deploying, and applying emerging digital technologies in computational biology to drive the next generation of therapeutic discovery. Perturbation screening is an essential approach for understanding how genetic and non-genetic changes impact cellular systems. By systematically probing gene function and network dependencies, these studies reveal disease mechanisms, identify novel therapeutic targets, and uncover biomarkers that can guide preclinical and translational research. This capability enables more confident target validation and accelerates the path from discovery to therapeutic hypothesis generation. In this role, you will develop and apply advanced computational and AI/ML methods to analyze perturbation screening data, strengthen data interpretability, and collaborate with experimental teams to ensure computational predictions translate into actionable insights for target and therapeutic discovery. The successful candidate will possess strong analytical aptitude, have a high degree of technical competency in computational sciences with a deep theoretical understanding of AI/ML and causal learning paired with an excellent understanding of molecular biology proven by a track record of innovative and collaborative research. RESPONSIBILITIES: Develop and apply advanced methodologies at the intersection of functional genomics and AI/ML analytics to identify and validate novel therapeutic targets, elucidate mechanisms of action, and discover preclinical biomarkers. Provide computational and data science support for high-throughput single-cell perturbation and phenotypic screening experiments (e.g., Perturb-seq) to map gene–phenotype relationships, uncover disease drivers, and prioritize hits for validation. Drive digital innovation by identifying gaps in analytical workflows and developing robust, reproducible, and FAIR solutions that improve data interpretability, strengthen readout confidence, and accelerate therapeutic hypothesis generation. Build and apply machine learning, deep learning, and generative AI approaches to simulate cellular responses to single or combinatorial genetic and non-genetic perturbations, delivering fit-for-purpose predictions that reduce experimental burden and accelerate therapeutic hypothesis generation. Collaborate closely with experimental biologists and translational teams to ensure computational predictions align with experimental design and biological context, enabling testable target and biomarker hypotheses; contribute to the design of perturbation experiments so that computational and experimental approaches inform and strengthen each other. What we expect of you We are all different, yet we all use our unique contributions to serve patients. The dynamic professional we seek is a senior data scientist with these qualifications. Basic Qualifications: Any degree and 8-13 years of directly related experience Preferred Qualifications: Demonstrated expertise in method development for functional genomics and perturbation screens, ideally with experience in single-cell omics technologies. Strong background in machine learning and AI, including deep learning, generative modeling, and transformer-based architectures; experience with pre-training, fine-tuning, and few-/zero-shot learning, ideally with experience in in silico modeling of cellular responses to drug or genetic perturbations. Proven track record of applying computational methods to drive biological insights, target validation, biomarker discovery, or therapeutic hypothesis generation. Proficiency in scientific programming languages and tool development using Python, R, or similar, with familiarity in relevant libraries and frameworks. Experience with large-scale data processing using cloud computing, workflow development, and software best practices (e.g., version control, continuous integration, test-driven development). Familiarity with agentic AI, digital innovation approaches, and FAIR data principles for building robust and scalable analytical workflows. Familiarity with molecular and disease biology, with the ability to contextualize computational findings in therapeutic discovery. Excellent analytical and communication skills, with the ability to extract and clearly present insights from complex data to diverse audiences with rigor and accuracy. Strong interpersonal and collaborative skills with demonstrated ability to thrive in cross-functional teams and effectively present results to diverse audiences. Creative, open-minded, and passionate about research, with a proven record of innovative algorithm and model development demonstrated through impactful publications, patents, or widely adopted tools. .