Jose G. Perez

Jose G. Perez

Applied ML Engineer & Applied Scientist

Ph.D. computer vision and scientific ML researcher who builds production-oriented ML systems: PyTorch models, Docker/Prefect infrastructure, CI/CD workflows, GPU-backed research pipelines, and full-stack engineering for long-running software products.

Download PDF Email
El Paso, Texas josegperezdev@gmail.com +1-915-780-8457 LinkedIn GitHub

Selected Impact

46.07%debris-covered ice IoU
28.2%relative improvement over benchmark
1.5 yrsML research engineering at JHU APL
1,000+AWBW full-stack commits

Experience

Machine Learning Ph.D. Intern

Johns Hopkins University Applied Physics Laboratory (APL)

May 2023 -- Dec 2024

  • Converted a multispectral satellite-image classification workflow into a Prefect-orchestrated backend service for scheduled and on-demand inference
  • Migrated model training from TensorFlow to PyTorch and trained triplet-loss models for real-time one-shot classification
  • Designed Docker/Docker Compose deployment workflows and GitLab CI/CD pipelines for linting, testing, and artifact packaging
  • Built data exploration interfaces and mission-dependency visualizations using OpenAI models and APL Dagger

Research Assistant

University of Texas at El Paso (UTEP)

May 2018 -- Dec 2025

  • Developed physics-guided glacier segmentation achieving 46.07% debris-covered ice IoU (28.2% over prior benchmark) via DEM-derived terrain augmentation and ITS_LIVE velocity loss
  • Built hybrid Physics-Informed LSTM for fluid-flow velocity prediction using Navier-Stokes constraints (73.7% U-vel, 85.3% V-vel improvement)
  • Built glacier velocity dataset pipeline fusing ITS_LIVE dynamics with Landsat/ICIMOD data across 7-year temporal aggregation; maintained GPU servers and ran 85+ controlled experiments
  • Built rat brain atlas alignment using OpenCV, SIFT, and RANSAC (published in Frontiers); additional work in Deep Q-Networks and simulation

Teaching Assistant

University of Texas at El Paso (UTEP)

Fall 2018 -- Summer 2025

  • Supported Data Structures, Computer Vision, Deep Learning, and Machine Learning at undergraduate and graduate levels
  • Delivered ML instruction for US Army learners at White Sands Missile Range

Undergraduate Research Assistant

University of Texas at Austin

June 2016 -- Aug 2016

  • Created multiscale model of T cell and antigen-presenting cell interactions using CompuCell3D in Python. Presented at 2016 BMES Conference.

Selected Projects

Physics-Guided Glacier Segmentation

PyTorch U-Net framework for Landsat 7 glacier segmentation with DEM physics channels, ITS_LIVE velocity constraints, 85+ experiment configs, and 46.07% debris-covered ice IoU.

Advance Wars by Web

Top contributor to a live multiplayer strategy game server with 1,000+ commits across PHP, Vue.js, Rust, MySQL, Docker, replay streaming, S3/MinIO storage, and UI/UX systems.

Physics-Informed LSTM

Hybrid LSTM+PINN architecture for 2D incompressible Navier-Stokes fluid-flow prediction, published at ASME FEDSM 2022.

Education

Ph.D. in Computer Science

University of Texas at El Paso, December 2025

Thesis: Physics-Guided Strategies for Enhancing Neural Networks Trained With Limited Data.

B.Sc. in Computer Science

University of Texas at El Paso, May 2018. Cum Laude. Minors in Biomedical Engineering and Mathematics.

Awards

Google-CAHSI Dissertation Award

Google / UTEP, 2022. $25,000.

BUILDing Scholars Traineeship

NIH-funded undergraduate research scholarship, 2015-2018.

Publications

  • Physics-Informed Glacier Ice Segmentation of HKH Region Using Multispectral Satellite Imagery. In progress, 2026. J. G. Perez, O. Fuentes.
  • Field Predictions of Hypersonic Cones Using Physics-Informed Neural Networks. ASME FEDSM, 2022.
  • Physics-Informed Long-Short Term Memory Neural Network Performance on Holloman High-Speed Test Track Sled Study. ASME FEDSM, 2022.
  • Empirical Game-Theoretic Methods to Minimize Regret Against Specific Opponents. SPIE Defense + Commercial Sensing, 2021.
  • Computer vision evidence supporting craniometric alignment of rat brain atlases. Frontiers in System Neuroscience, 2018.

Skills

Languages Python, Java, TypeScript, Rust, C#, SQL
ML & CV PyTorch, PyTorch Lightning, OpenCV, scikit-learn, Numba, U-Net, PINNs, triplet loss
MLOps & Infra Prefect, Docker, Docker Compose, GitLab CI/CD, Linux, CUDA, GPU servers
Domains Computer vision, multispectral imagery, satellite imagery, geospatial ML, scientific ML, CFD