- Pay
- $77–126k
- Where
- Golden, CO · Remote Site · Remote
- Level
- Mid
Posting Title
Postdoctoral Researcher in uncertainty quantification and prediction reliability .
Location
CO - Golden .
Position Type
Postdoc (Fixed Term) .
Hours Per Week
40 .
Working at NLR
NLR is located at the foothills of the Rocky Mountains in Golden, Colorado is the nation's primary laboratory for energy systems research and development.
Join the National Laboratory of the Rockies (NLR), where world-class scientists, engineers, and experts are accelerating energy innovation through breakthrough research and systems integration. From our mission to our collaborative culture, NLR stands out in the research community for its commitment to an affordable and secure energy future. Spanning foundational science to applied systems engineering and analysis, we focus on solving complex challenges to deliver advanced, secure, reliable, and cost-effective energy solutions. Our work helps strengthen U.S. industries, support job creation, and promote national economic growth.
At NLR, you'll find a mission-driven environment supported by state-of-the-art facilities, multidisciplinary research teams, and strong collaborations with industry, academia, and other national laboratories. We offer robust professional development opportunities, and a competitive benefits package designed to support your career and well-being.
Job Description
The Artificial Learning and Intelligent Systems group in NLR’s Computational Science Center has an immediate opening for a Postdoctoral Researcher to pursue a cross-disciplinary research project in uncertainty quantification, with particular emphasis on evaluating prediction reliability across scales (lab-scale to pilot-scale) in applications relevant to the DOE Office of Critical Minerals and Energy Innovation. We are looking for an individual who can reason about uncertainty arising from sparse data, extrapolation, and potentially incorrect model structure.
The successful applicant will develop methods to quantify uncertainty in cross-scale prediction models and identify when a model is operating beyond the conditions where its predictions are reliable. The deeper problem is that a model trained at one scale may be systematically wrong at another because the underlying relationships have changed, important mechanisms are missing, or the available evidence does not adequately constrain the prediction. The successful applicant will conduct original methods research to address this challenge. We are looking for individuals with background in applied statistics, uncertainty quantification for scientific computing, statistical learning theory, reliable or trustworthy machine learning, out-of-distribution detection and distributionally robust methods, and/or model validation and discrepancy modeling.
.
Basic Qualifications
Must be a recent PhD graduate within the last three years.
* Must meet educational requirements prior to employment start date.
Additional Required Qualifications
Demonstrated experience with statistical modeling including strong grounding in statistical inference, Bayesian methods, probabilistic modeling, and uncertainty propagation.
Demonstrated experience with uncertainty quantification including experience distinguishing and estimating aleatoric and epistemic uncertainty, and model-form or structural uncertainty.
Demonstrated understanding of methods for assessing whether predicted uncertainty corresponds to actual prediction errors, especially with limited validation data.
Demonstrated understanding of distribution shift and out-of-distribution detection for identifying when new conditions or data differ meaningfully from the training domain.
Demonstrated experience with model discrepancy and structural uncertainty to enable understanding how missing physics, incorrect assumptions, or changing mechanisms affect model reliability.
Demonstrated experience evaluating predictions outside the training domain, including methods that can recognize when extrapolation is unsafe or…