Senior Data Scientist III - AI Assurance Researcher

Pacific Northwest National Laboratory•Richland, WA

15h•Onsite

About The Position

Pacific Northwest National Laboratory (PNNL) is seeking a deep learning researcher in its AI and Data Analytics Division within the National Security Directorate. The role investigates how AI models learn and make decisions, and develops methods to assess and improve model reliability. The researcher connects internal representations and computations to observable behavior to explain model decisions and assess how uncertainty, changing conditions, or adversarial inputs affect output reliability. The research addresses national security and safety needs, with opportunities to collaborate across the U.S. government and advance AI Assurance research. The position leads projects or major components of larger efforts, develops research questions and methods, and mentors colleagues while remaining hands-on in experimental research. Work may include deep learning fundamentals such as optimization, representations, and generalization; large-scale experiments using high-performance computing; and research involving large language and vision models. The position is based in Richland, Washington, or Seattle, Washington, and requires an onsite presence. It requires the ability to obtain and maintain a federal security clearance, U.S. citizenship, and compliance with applicable background investigation and drug-testing requirements.

Requirements

BS/BA and at least 5 years of data science experience, or MS/MA and at least 3 years of data science experience, or PhD and at least 1 year of data science experience.
Experience conducting deep learning research to investigate how models learn, behave, or generalize.
Experience using Python and PyTorch for deep learning research.
U.S. citizenship.
Ability to obtain and maintain a federal security clearance and meet eligibility requirements for access to classified matter.
Must meet the stated drug-testing and illegal-drug non-use requirements for security clearance processing.

Nice To Haves

Experience explaining model behavior through analysis of internal representations and computations, including mechanistic interpretability of large language or vision models.
Experience developing or evaluating methods for assessing tool-using models and multi-step agentic workflows, including analysis of execution traces, task outcomes, and failure modes.
Experience designing and executing large-scale deep learning experiments.
Experience leading research projects or major technical efforts, including formulating research questions, developing or adapting methods, coordinating contributors, and evaluating and communicating results.
Experience translating research prototypes into tools or capabilities, or contributing peer-reviewed publications or other technical contributions in relevant research areas.
Experience supporting national security missions.

Responsibilities

Lead research projects or major components of larger efforts, coordinating technical approaches, team contributions, and milestones.
Formulate research questions, develop or adapt methods, and design and execute deep learning experiments, including large-scale experiments on high-performance computing infrastructure as appropriate.
Investigate how model architectures, training dynamics, and internal computations shape behavior, including capabilities and failure modes of large language and vision models.
Develop methods to quantify uncertainty and assess robustness under distribution shifts or adversarial inputs; use red and blue teaming to identify vulnerabilities and evaluate defensive methods.
Develop reproducible evaluations to examine model reliability and failure modes, compare methods against appropriate baselines, and communicate findings, uncertainty, and limitations.
Develop reproducible research code and collaborate with engineers to translate findings and prototypes into useful tools and capabilities.
Lead preparation of publications, technical reports, and presentations communicating research findings.
Mentor researchers and engineers through technical discussions, feedback, and collaborative research.
Develop research ideas and contribute to proposals with collaborators.
Conduct work in secure environments in accordance with operational security requirements.