CV

Ph.D.

University of New Mexico, School of Engineering

Computer Science

• Dissertation: Seeking Structure in Complex Systems: From Feature Analysis to Space-Time Causal Discovery with Earth Science Applications
• Advisor: Melanie E. Moses

M.B.A.

University of New Mexico, Anderson School of Management

Business Administration

B.S.

University of New Mexico, School of Engineering

Computer Science

Senior Member of Technical Staff (R&D S&E, Artificial Intelligence)

Sandia National Laboratories

AI Credibility department.

• Leading the credibility team on the Towards Certifying Trustworthy Machine Learning LDRD, including development of the Generalized Credibility Maturity Model (GCMM).
• Causal modeling for cyber-physical threat detection in power grids (MetaShield LDRD) and causal analysis of weather foundation models (SEA-CROGS).

Postdoctoral Appointee

Sandia National Laboratories

Scientific Machine Learning department.

• Research on AI credibility: causal discovery, trustworthy and certifiable machine learning, and physics-informed modeling for high-consequence applications.
• Causal analysis of weather foundation models to test whether systematic interventions bias model behavior toward more (or less) realistic weather phenomena.
• Causal modeling for cyber-physical threat detection in power grids.

R&D Graduate Intern (Year-Round)

Sandia National Laboratories

Scientific Machine Learning department. Part-time 2019-2021, full-time 2021-2025.

• Developed CaStLe, a causal discovery meta-algorithm for recovering local space-time causal structure in gridded data.
• Led causal modeling for the CLDERA Grand Challenge LDRD, identifying source-to-impact pathways in the climate system.
• Applied and benchmarked causal discovery (PC, PCMCI) for causal pathways from the 1991 Mt. Pinatubo eruption.
• Used random forest feature importance to compare observed and simulated climate data and expose model deficiencies.

Graduate Research Assistant

University of New Mexico

Department of Computer Science, School of Engineering.

• Robotics research under Dr. Lydia Tapia.
• Computational sociology research under Dr. Marina Kogan.

Owner

Swarming Technologies LLC

Profitable robotics company selling self-designed autonomous robots, custom-design consulting, and repair services.

• Robots, known as 'Swarmies', were designed for swarm robotics and autonomous robotics research and education.
• Swarmies were featured in the NASA Swarmathon, NASA MINDS competition, and CS4All NM.

Robot Engineer/Designer

NASA Swarmathon & Moses Biological Computation Lab

Designed and built the autonomous swarming robots used in the NASA Swarmathon, a nation-wide collegiate robotics competition.

• Mechanical design in Autodesk Inventor; manufacturing via SLS, SLA, and FDM 3D printing, plus electronics development.
• Analyzed competition data to track progress and success determinants using MySQL and Python (NumPy, SciPy, Pandas).
• Automated code deployment to robot fleets using Ansible and Docker.
• Swarm robotics research on Arduino/iPod Touch-controlled iAnt robots, tuning behavior with a genetic algorithm.

Software Engineering Intern

Intel Corporation

Product surveying, QA, troubleshooting, and testing.

• Arduino/Galileo programming and debugging.
• Designed and built a robotic car controlled by the Intel Galileo Gen 2, with Bluetooth input from an Android phone and sensors for line following and obstacle avoidance.

MetaShield LDRD: Co-Lead, Causal AI Team

PI: Georgios Fragkos. Combining causal AI with meta-learning to enhance ML detection of cyber-physical threats to power grids.

• Co-leading the causal AI team to integrate causal modeling paradigms with ML for better power grid analysis and resilience.
• Cross-lab team spanning centers 1400, 5600, 8700, and 8800, with industry partners PNM and GE Vernova.

SEA-CROGS: Contributor

PI: Eric Cyr. Scalable, Efficient and Accelerated Causal Reasoning Operators, Graphs and Spikes for Earth and Embedded Systems.

• Developing causal analyses of the Aurora weather foundation model to determine whether systematic interventions can bias model behavior toward more (or less) realistic weather phenomena.
• Collaboration between PNNL and SNL, with academic partners at Brown, Yale, Caltech, NJIT, and Stanford.

Towards Certifying Trustworthy Machine Learning LDRD: Team Lead, Credibility

PI: Michael Darling. Leading the credibility assessment effort.

• Developing the Generalized Credibility Maturity Model (GCMM), which generalizes the CompSim Predictive Capability Maturity Model (PCMM) to any modeling paradigm.
• Developing the taxonomy of computational modeling.
• Cross-lab team spanning centers 1400, 5400, 5500, and 8700.

CLDERA Grand Challenge LDRD: Causal Modeling Lead, Attribution Thrust

PI: Diana Bull; Thrust Lead: Laura P. Swiler. CLimate Impact: Determining Etiology thRough pAthways.

• Led the R&D of applying causal modeling to understand the drivers of stratospheric aerosol transport and their impacts.
• Led the R&D of a novel causal discovery algorithm (CaStLe) for gridded spatiotemporal data in physical systems.

Causal Evaluations for Identifying Differences Between Observations and Earth System Models (Late-Start LDRD): Contributor

PI: Matt Peterson.

• Developed a causal discovery methodology to model Arctic sea ice extent drivers, in preparation for applying the workflow in the CLDERA Grand Challenge LDRD.

Advanced Data Analytics for Proliferation Detection (ADAPD): Contributor

A multi-lab initiative.

• Developed hidden Markov models for multimodal temporal data analysis.

Arctic Tipping Points Triggering Global Change LDRD: Contributor

PI: Kara Peterson.

• Developed random forest regression models for predicting Arctic sea ice extent with feature importance analysis.

M-CaStLe: Uncovering Local Causal Structures in Multivariate Space-Time Gridded Data

In preparation, target: Transactions on Machine Learning Research (TMLR)

Causal Graphs for Threat Detection in Cyber-Physical Power Grids

Submitted to IEEE Resilience Week (RWS) 2026 (under review)

Tracing the Space-Time Causal Origins of Earth System Extremes

Submitted to Science Advances (under review)

Causal AI-Powered Dynamic ML Model Orchestration for Robust Cybersecurity

IEEE SmartGridComm 2026 (to appear)

A Multi-Objective Optimization Framework for Trustworthy Machine Learning

Modeling Decisions for Artificial Intelligence (MDAI) 2026 (to appear)

Challenges of Certifying Machine Learning Trustworthiness in High-Consequence Domains

Data Science in Science (accepted)

From Incredible Claims to Credible AI: Evidence-Backed Maturity Profiles for Risk-Informed AI Assurance

AAAI Fall Symposium on AI Trustworthiness and Risk Assessment for Challenged Contexts (ATRACC) 2026 (accepted)

Seeking Structure in Complex Systems: From Feature Analysis to Space-Time Causal Discovery With Earth Science Applications

PhD dissertation, The University of New Mexico

Space-Time Causal Discovery in Earth System Science: A Local Stencil Learning Approach

Journal of Geophysical Research: Machine Learning and Computation

Introduces Causal Space-Time Stencil Learning (CaStLe), a meta-algorithm that leverages regularities in local space-time dependencies to learn governing global dynamics. This local perspective eliminates spurious confounding and drastically reduces sample complexity, making space-time causal discovery practical and effective, including for non-periodic, transient phenomena such as volcanic eruption plumes.

CLimate Impact: Determining Etiology thRough pAthways (CLDERA)

Sandia National Laboratories, Tech. Rep. SAND2024-13423R

Benchmarking the PCMCI Causal Discovery Algorithm for Spatiotemporal Systems

Sandia National Laboratories technical report

Benchmarks the PCMCI causal discovery algorithm on gridded spatiotemporal systems, finding that it requires unrealistic sample sizes and suffers a notable curse of dimensionality, which motivated the development of CaStLe.

Water Cycle-Driven Infectious Diseases as Multiscale, Reliable, Continuously Updating Water Cycle Sensors

Sandia National Laboratories white paper

Advancing Sea Ice Predictability in E3SM with Machine Learning

Sandia National Laboratories white paper

Causal Evaluations for Identifying Differences between Observations and Earth System Models

Sandia National Laboratories technical report

Learning Why: Data-Driven Causal Evaluations of Climate Models

ICML 2021 Workshop: Tackling Climate Change with Machine Learning

Machine Learning Feature Analysis Illuminates Disparity Between E3SM Climate Models and Observed Climate Change

Journal of Computational and Applied Mathematics

Uses random forest regression and Gini importance to show that the Energy Exascale Earth System Model (E3SM) relies too heavily on one of ten climatological quantities to predict September sea ice averages, and overweights six others relative to observed data.

Arctic Tipping Points Triggering Global Change (LDRD Final Report)

Sandia National Laboratories technical report

The Swarmathon: An Autonomous Swarm Robotics Competition

ICRA 2018 Workshop: Swarms — From Biology to Robotics and Back

Best Talk Prize

European Seminar on Computing (ESCO)

3rd Place Poster Prize

Department of Energy Conference on Data Analysis (CoDA)

Reviewer

IEEE Transactions on Neural Networks and Learning Systems

Peer review for IEEE TNNLS.

Reviewer

American Meteorological Society, Journal of Climate

Peer review for the Journal of Climate.

Reviewer

ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Peer review for KDD.

Adaptive Ski Instructor

Adaptive Sports Program New Mexico Inc

Teach skiing to people with various disabilities.

Troop 9 Board of Review Member

Boy Scouts of America

Attend board of review meetings to assist in scout advancement.