Senior Data Scientist/PI - Evaluating Foundation Models and SI Agents for National Security

Overview At PNNL, our core capabilities are divided among major departments that we refer to as Directorates within the Lab, focused on a specific area of scientific research or other function, with its own leadership team and dedicated budget. Our Science & Technology directorates include National Security, Integrated Discovery Sciences, and Energy and Environment. In addition, we have an Environmental Molecular Sciences Laboratory, a Department of Energy, Office of Science user facility housed on the PNNL campus. The National Security Directorate (NSD) drives science-based, mission-focused solutions to take on complex, real-world threats to our nation and the world. The AI and Data Analytics Division, part of NSD, combines profound domain expertise and creative integration of advanced hardware and software to deliver computational solutions that address complex data and analytic challenges. Working in multidisciplinary teams, we connect foundational research to engineering to operations, providing the tools to innovate quickly and field results faster. Our strengths are integrated across the data analytics lifecycle, from data acquisition and management to analysis and decision support. Read more about the AI & Data Analytics division: https://www.pnnl.gov/ai-and-data-analytics Responsibilities We are seeking a Senior Data Scientist to lead cutting-edge applied research for national-security applications. In this role, you will move beyond traditional evaluation practices and aggregate benchmark scores to design comprehensive evaluation harnesses, lead novel test & evaluation (T&E) method development, and help shape the technical direction of T&E for large programs. As a Senior Data Scientist, you will own critical evaluation workstreams and lead small to medium research projects, translate national security needs into novel evaluation frameworks, and champion rigorous, reproducible data science and research practices. In addition to technical execution, you will actively mentor junior staff and lead business development and proposal efforts to secure funding that aligns with or expands our division’s AI research and T&E capabilities. Key Responsibilities: • Project & Technical Leadership: Own defined workstreams or small-to-medium projects evaluating LLMs, VLMs, AI agents, agentic workflows, and integrated AI systems for national-security missions. • Technical Leadership: Translate complex national security mission needs into strategic T&E frameworks, operational scenarios, evaluation criteria, and technical approaches specifically tailored for generative AI (large language models (LLMs), vision-language models (VLMs)) and agentic pipelines. • Mentorship: Mentor junior staff and interns and coordinate work across multidisciplinary teams to grow junior staff into technical leaders. • Evaluation Framework Design: Assess systems at the model, component, workflow, and end-to-end levels. Design evaluation plans that integrate aggregate performance measures with behavioral analysis, failure characterization, meaningful data slices, robustness testing, calibration and uncertainty, and mission-context evidence. • Harness & Tool Development: Select, design, and/or develop evaluation environments, test harnesses, agentic harnesses, datasets, deterministic baselines, rubrics, and validated automated or model-based graders. • Technical & Collaborative Leadership: Serve as a role model for research and engineering best practices within the team. Champion collaborative data science workflows, ensuring all research, code, and models are version-controlled, fully documented, and reproducible across your project team(s) using our collaborative tools (Git, Gitlab, Jira, Confluence, Microsoft 365, etc.). • Technical Communication: Synthesize evaluation results, operational constraints, and uncertainties into technical reports, documentation, and peer-reviewed publications. Present research at technical conferences through accepted papers or invited talks. • Stakeholder Engagement: Deliver compelling briefings tailored to both technical experts and non-technical stakeholders, clearly connecting technical findings to mission impact. Drive technical discussions and deliver briefings to sponsors and domain experts to inform and shape project direction. Work Location & Requirements: • This position is based in Richland, WA or Seattle, WA and requires an onsite presence. • This position requires the ability to obtain and maintain a federal security clearance. (see below for details) Qualifications Minimum Qualifications: • BS/BA and 5+ years of data science experience -OR- • MS/MA and 3+ years of data science experience -OR- • PhD with 1+ year of data science experience • Experience setting technical direction and leading technical work of yourself and others on research projects. Preferred Qualifications: • Experience as a principal investigator of federally funded research and development projects, or owning an evaluation workflow or small-to-medium technical project (leading technical work of yourself and others) involving LLMs, VLMs, or AI agents. • Experience developing and leading the design of evaluation environments, agentic harnesses, behavioral tests, validated graders, or continuous-evaluation pipelines (including "LLM-as-a-judge" methods or evaluation agents). • Experience analyzing model behavior beyond aggregate performance metrics, including failure modes, data-slice performance, robustness, generalization, calibration, uncertainty, or distribution shift. • Experience evaluating systems that combine generative AI with deterministic NLP, computer vision, retrieval, knowledge graphs, rules, or analytical tools. • Strong experience using Python and frameworks such as PyTorch to conduct reproducible machine-learning research. • Robust experience with version control (Git) and collaborative software development practices to maintain reproducible research repositories. • Evidence of