Geospatial Data Engineer

Oak Ridge National Laboratory•Oak Ridge, TN

26d•Onsite

About The Position

The Geospatial Data Modelling Group within the Human Dynamics Section, part of the Geospatial Science and Human Security Division at ORNL, is seeking a Geospatial Data Engineer to support research and operational workflows focused on scalable geospatial data science, applied machine learning, and production-grade engineering practices to deliver repeatable, defensible, and time-dynamic geospatial products in support of national security, humanitarian response, disaster assessment, and resilience planning. In this technical role, the candidate will collaborate with an interdisciplinary team of human geographers, population scientists, geospatial analysts, data scientists, and software engineers. They will contribute across the full lifecycle of geospatial modeling efforts: data acquisition and preparation, feature engineering, model development and evaluation, MLOps and codebase maintenance, automation, and quality assurance. A key component of this position is building agentic AI workflows that help discover, gather, validate, and standardize open-source data for downstream geospatial analytics and machine learning. The position offers a unique opportunity to work on applied spatial analytics and geospatial data modeling at scale, leveraging diverse geospatial, demographic, and remotely sensed data sources. While the role does not require independent development of novel AI algorithms, it does require strong implementation skills, sound statistical judgment, and an ability to translate methods into reliable, maintainable, and well-documented pipelines.

Requirements

Bachelor’s degree and 3+ year’s experience in Geography, GIScience, Computer Science, Data Science, Statistics, Engineering, or a related field with a strong quantitative and software development emphasis.
Demonstrated experience with geospatial analysis using Python in a production or research to production environment leveraging common geospatial libraries (e.g., geopandas, rasterio, shapely, pyproj) and/or enterprise GIS tooling (e.g., PostGIS).
Strong software engineering fundamentals: Git-based workflows, testing, code review, and writing maintainable, well-documented code.
Experience preparing and validating raster and vector datasets (data cleaning, transformation, projection/CRS management, and quality control).
Working knowledge of machine learning and statistical modeling concepts (e.g., regression, classification, clustering, model evaluation).
Ability to work effectively in a team-based, production-oriented research environment and communicate technical results to diverse stakeholders.

Nice To Haves

Master’s degree in a relevant discipline or equivalent applied experience in geospatial data science, MLOps, or applied machine learning.
Experience with modern MLOps tooling and practices (e.g., MLflow or equivalent experiment tracking, model registries, containerization, reproducible environments).
Experience building data pipelines and workflow orchestration (e.g., Airflow, Prefect, Dagster, Make/Snakemake) and working in Linux/HPC environments.
Experience with large, multi-resolution geospatial datasets and performance-oriented processing (tiling, chunking, parallelization; Dask/Spark a plus).
Experience using or building agentic/LLM-enabled workflows for data discovery, extraction, and normalization, with attention to provenance, reproducibility, and quality.
Familiarity with uncertainty, data limitations, and bias in population and demographic modeling and in applied geospatial decision-support contexts.
Active or eligible U.S. security clearance or ability to obtain one.

Responsibilities

Develop, maintain, and operationalize geospatial data science pipelines across ingestion, feature engineering, training, inference, evaluation, and delivery, using reproducible MLOps practices (version control, testing, experiment tracking, containerization, and CI/CD).
Support implementation of agentic AI workflows to discover, gather, and prepare data from open-source repositories (e.g., catalogs, APIs, and bulk downloads), including provenance tracking, metadata extraction, and licensing/usage notes.
Build scalable geospatial data preparation and validation routines for raster and vector data (projection harmonization, spatial joins, tiling/chunking, and QA/QC).
Develop geospatial validation frameworks for model outputs (e.g., comparisons to reference datasets, spatial cross-validation, summary dashboards, and automated report generation).
Support documentation, metadata development, and version tracking for data products and model releases; contribute to technical summaries, figures, and reports/publications as appropriate.
Participate in code reviews, model reviews, and data readiness reviews to ensure analytical defensibility, transparency, and fitness-for-use in operational and decision-support contexts.
Collaborate with research staff to integrate new data sources, indicators, and modeling approaches into existing workflows; communicate clearly across technical and domain teams.