Skip to content
View brianurban's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report brianurban

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
brianurban/README.md

Hi, I'm Brian! πŸ‘‹

Data science professional bridging physics domain expertise with machine learning engineering. MSc in Data Science (DePaul) and BSc in Physics with Applied Mathematics (NEIU). Built predictive models, deployed cloud-based analytics apps, and managed QA data across 700+ assets at Argonne National Laboratory. Currently seeking data science roles where interdisciplinary problem-solving drives measurable impact.

πŸŽ“ MSc Data Science (DePaul, 2024–2026) | BSc Physics + Applied Mathematics (NEIU, 2015–2021)

πŸ”¬ Research Interests: Deep learning applications in astrophysics: NEO hazard classification, nuclear fusion prediction in stars, and model robustness under measurement uncertainty.

πŸ”­ What's Next:

  • Actively interviewing for data science / ML engineering roles
  • Exploring PhD programs in astrophysics (data science Γ— nuclear astrophysics)
  • Expanding horizons at the intersection of astrophysics and data science

πŸš€ Featured Projects

  • β˜„οΈ Comparative ML for NEO Hazard Classification | Jan – Mar 2026
    Built an asteroid hazard classifier using SMOTE and gap-penalized loss on a highly imbalanced dataset (12.7% positive rate). Validated engineered orbital features via K-Means clustering, achieving 2.3Γ— enrichment of hazardous classes.

  • 🌌 Galaxy Morphology Classifier | Sept – Nov 2025
    Fine-tuned an EfficientNet-B0 CNN for 10-class galaxy classification on 17,736 DECaLS images using transfer learning and weighted loss, reaching 83% test accuracy vs. 15% for classical baselines.

  • ✨ Stellar Classification Comparative Analysis | Apr – Jun 2025
    Compared MLP, logistic regression, and decision trees across six Gaussian noise levels using 5-fold CV. Found that peak clean-data accuracy is a misleading model selection metric for noisy scientific data.


πŸ› οΈ Tech Stack

Languages: Python R SQL Markdown LaTeX HTML CSS JavaScript PowerShell

Libraries: Pandas NumPy PyTorch scikit-learn Matplotlib Seaborn SciPy

Tools: Positron Visual Studio Code Jupyter RStudio Tableau Docker AWS Microsoft Azure React Git


πŸ’Ό Experience

  • πŸ“š Math & Physics Tutor | Malcolm X College | Nov 2024 – Present
    Developed data-informed learning frameworks, delivering workshops on quantitative and statistical methods to students of varying technical fluency.

  • βš™οΈ Electro-Mechanical Survey Technician | Argonne National Laboratory | Feb – Sept 2023
    Managed QA data for 700+ magnet assemblies on the Electron-Ion Collider upgrade project, applying physics-informed diagnostics and building structured data pipelines for cross-team traceability.

  • πŸ’» Remote Full-Stack Software Developer | Revature | May – Sept 2022
    Built predictive analytics modules and interactive data dashboards, deploying scalable Azure cloud solutions and React-based visualization interfaces for real-time decision-making.


πŸ“« Let's Connect

LinkedIn GitHub


⭐️ Beyond the Code

  • I share my passion and understanding of the night sky by bringing my telescope, which I use for astrophotography, to local star parties.
  • I enjoy playing the piano, singing in a choir, doing art, and cooking.
  • When I'm not stargazing, you'll find me playing Metal Gear Solid, Kingdom Hearts, and Final Fantasy.

Popular repositories Loading

  1. stellar-classification-comparative-analysis stellar-classification-comparative-analysis Public

    Comparative analysis of ML model robustness to measurement noise on stellar classification using PyTorch, scikit-learn, and 5-fold cross-validation.

    Jupyter Notebook

  2. brianurban brianurban Public

  3. neo-hazard-classification neo-hazard-classification Public

    Comparative analysis of 12 machine learning models (deep learning + classical ML + ensembles) for near-Earth object (NEO) hazard classification on imbalanced orbital data (338K records, 6.84:1 ratio).

    Jupyter Notebook

  4. galaxy-morphology-classification galaxy-morphology-classification Public

    Comparative study of EfficientNet-B0 transfer learning vs. traditional ML (SVM, RandomForest, ExtraTrees, kNN) on 28 hand-crafted morphological features for galaxy classification using the Galaxy10…

    Jupyter Notebook