Skip to content
View hamdanashfaq63's full-sized avatar
  • San Diego State University
  • San Diego, CA

Block or report hamdanashfaq63

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.

Content in all repositories owned by your account will be closed.
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
hamdanashfaq63/README.md

Hamdan Ashfaq

PhD student in Computational Science at UC Irvine and San Diego State University. I build machine learning methods for materials characterization, mostly electron microscopy: finding hidden order in 4D-STEM diffraction data and classifying microstructures from SEM images.

Advised by Prof. Mehmet Dogan (SDSU) and Prof. Ruqian Wu (UCI). Based in San Diego.

Research

The Ohio State University, Center for Electron Microscopy and Analysis (CEMAS) · Visiting Fellow, Hwang Lab

  • 4D-STEM multislice simulations in collaboration with Sandia National Laboratories.
  • Self-supervised learning on 4D-STEM nanodiffraction patterns of amorphous semiconductors (Intel-funded SURE fellowship). Poster at Microscopy & Microanalysis 2024.
  • Co-author on a manuscript under review at Acta Materialia.

Purdue University · Summer Undergraduate Research Fellow, Chawla group

  • ResNet-50 and U-Net++ models that classify Sn-58Bi solder microstructures from SEM images with 90.1% accuracy.

Denison University · Senior capstone

  • Automated fold symmetry detection across 65,536 nanodiffraction patterns of CuZr metallic glass. Best Senior Capstone Project, Department of Data Analytics.

Projects

Project What it does Stack
Finding Order in Disorder Detects local 2- to 6-fold symmetry in 4D-STEM patterns of metallic glass, with synthetic validation and a characterization of where rule-based methods break Python, scikit-image, NumPy, pytest
PJM Load Forecast Accuracy Client project for AEP Energy measuring how PJM demand forecast error changes with horizon, time of week and weather R, tidyverse, Python, pandas
FPN Delivery Team-built mobile app that connects homebound food pantry clients with volunteer drivers, with live delivery tracking, SMS sign in and an admin dashboard React Native, Expo, Supabase

Tools

  • ML: PyTorch, TensorFlow, CUDA, multi-GPU training, scikit-learn, CNNs (ResNet, U-Net++), self-supervised learning (SimSiam)
  • Scientific computing: NumPy, SciPy, scikit-image, 4D-STEM simulation, HPC at the Ohio Supercomputer Center
  • Data and apps: pandas, R and the tidyverse, React Native, Supabase

Education

  • PhD, Computational Science, SDSU and UC Irvine Joint Doctoral Program, 2026 to present
  • Computer Science and Data Analytics, Denison University, 2026. Denison Phi Society (top 7% of class)

Pinned Loading

  1. DA401 DA401 Public

    Detecting local fold symmetry in 65,536 4D-STEM nanodiffraction patterns of CuZr metallic glass. Python package with tests and synthetic validation.

    Jupyter Notebook

  2. AEP_Energy_Project AEP_Energy_Project Public

    How accurate are PJM electricity demand forecasts? Error by horizon, time of week and weather, for AEP Energy. R and Python.

    R

  3. volunteer-delivery-app volunteer-delivery-app Public

    Volunteer food delivery app for the Licking County Food Pantry Network. Expo, React Native and Supabase. CS 349 team project.

    JavaScript