I lead end-to-end machine learning systems at the National Laboratory of the Rockies (NREL), where my models power the DOE Section 179D Web Calculator—a nationally deployed, IRS-supported platform. I trained ridge regression and deep neural network surrogates on EnergyPlus simulations across 16 U.S. climate zones, achieving 1.1–3.3% average CVRMSE.

Beyond the 179D platform, I've built LangGraph-based LLM agents, and co-developed a two-stage deep reinforcement learning framework combining evolution strategies and PPO for grid-interactive building control. My work spans 20+ peer-reviewed publications (401 citations, h-index 11), one U.S. patent, and production MLOps pipelines using MLflow, DVC, Evidently, Prometheus, Grafana, GitHub Actions, AWS SageMaker, and Azure ML.

National Laboratory of the Rockies (NREL) · Researcher, Residential Buildings Group Jun 2020 – Apr 2026 Led end-to-end development of production-grade ML systems for the DOE Section 179D Web Calculator, owning data engineering, model development, validation, deployment, and monitoring. Built LangGraph and LLM-based workflows, with deterministic evaluation against simulation ground truth with acceptance criteria including RMSE ≤ 0.5°C and parameter CV ≤ 0.15. Developed HVAC fault-detection methods using thermostat drive-cycle data and grey-box thermal models, published in Energy and Buildings (2021) and Applied Thermal Engineering (2024).

National Laboratory of the Rockies (NREL) · Postdoctoral Researcher, Residential Buildings Group Jan 2019 – May 2020 Developed research methods for building energy modeling, system identification, HVAC fault detection, forecasting, and optimization. Applied extended Kalman filtering and grey-box thermal models to noisy operational measurements and generated simulation datasets across building configurations, equipment, and climate conditions.

Opera Solutions · Senior Analytics Specialist Jun 2018 – Dec 2018 Directed a data analytics team developing statistical and machine learning models to forecast used-vehicle floor prices at auction. Created model validation and performance-reporting workflows, and translated forecast results into pricing and bidding recommendations for business stakeholders.

SKILLS WANTED: Python, PyTorch, TensorFlow, scikit-learn, LangGraph, FastAPI, Apache Spark, SQL, C, C++, MATLAB, PySpark · Deep learning, LLMs, reinforcement learning (PPO, evolution strategies), time-series forecasting, MLOps, explainable AI (SHAP), computer vision (semantic segmentation, OCR), physics-informed ML, optimization (MILP, MPC), feature engineering, hyperparameter tuning, A/B testing, transfer learning · MLflow, DVC, Evidently, Docker, Kubernetes, CI/CD, GitHub Actions, AWS SageMaker, Azure ML, Prometheus, Grafana, EnergyPlus, OpenStudio, ResStock, IfcOpenShell

EDUCATION: Ph.D. Mechanical Engineering, Texas A&M University, 2018 · M.S. Mechanical Engineering, Texas A&M University, 2011