Research

Modified

September 1, 2026

Research Statement

Data Management: Frameworks

Much of applied research fails not from a lack of data, but from a lack of structure for managing it. This line of work focuses on designing and implementing research data management frameworks that follow FAIR principles (Findable, Accessible, Interoperable, Reusable), with particular attention to three things:

  • Data governance: defining roles, permissions, and responsibilities across the data lifecycle, from field collection through long-term archiving.
  • Metadata & traceability: using standardized metadata schemas (e.g., Dublin Core) so a third party can understand and reuse a dataset without additional context.
  • Reproducibility: versioned workflows (Git, containers, declarative environments) that guarantee an analysis can be exactly repeated, even years later.

Generative AI: Tools for Research

Language models are changing how science gets done, but their value depends on using them with judgment. This line explores concrete applications and their limits:

  • Assisted literature review: screening and initial synthesis of large volumes of papers, always with human verification of cited sources.
  • Exploratory data analysis: generating reproducible code (Python/R) for cleaning, visualization, and preliminary modeling — speeding up the exploratory phase without replacing rigorous statistical analysis.
  • Synthetic data: generating simulated data to test models or protect sensitive information before sharing a dataset. Ethical & methodological limits: transparently documenting where and how generative AI was used in a project, avoiding treating it as an unverified source of truth.

Workforce Development: Technology Transfer

Research and new technology only create value once people can actually use them. This line focuses on workforce development — building the training, documentation, and skill pathways that let research outcomes move out of the lab and into everyday practice:

  • Technical training programs: structured courses and workshops that bring practitioners up to speed on new tools, methods, or technologies coming out of the research.
  • Documentation & onboarding materials: clear, practice-oriented guides (not academic papers) that let a new user adopt a tool or method without direct supervision.
  • Skills-gap analysis: identifying where a workforce’s current capabilities fall short of what a new technology or method requires, and designing training to close that gap.
  • Industry & institutional partnerships: working directly with employers, agencies, or training bodies so that technology transfer includes the human capacity to sustain it, not just the technology itself.

Funding

  • AI Institute: Agricultural AI for Transforming Workforce and Decision Support (AgAID), USDA-NIFA, in partnership with NSF, Award No.2021-67021-35344.
  • Washington Soil Health Initiative supported by the State of Washington and administered by WSCC, WSDA, and WSU.