Dr. Shahadat Hussain — Computational Materials Science & Data-Driven Manufacturing

// about

The lab bench is a terminal.

Modern materials engineering lives at the intersection of geometry, physics, and data. My computational work spans the full loop: generating triply periodic minimal surface (TPMS) fields, discretizing them into printable meshes, simulating their mechanical response, and closing the loop with machine-learning models trained on real laser powder bed fusion (LPBF) process data.

This approach lets me do things that are impractical experimentally — sweep hundreds of lattice parameters overnight, predict where balling or spattering will occur before a single laser fires, and quantify microstructural features directly from SEM imagery instead of eyeballing them.

For industry partners in aerospace and defense, this translates to three things: faster iteration, lighter parts, and qualified processes.

6+years_r&d_experience

3lattice_topologies_mastered

10+publications & datasets

∞process_parameter_sweeps

// tech_stack

Tools of the trade

A scientific-computing stack purpose-built for computational materials design and manufacturing analytics.

language / core

Python (NumPy, SciPy, pandas)
MATLAB / numerical methods
C++ for performance-critical kernels
Bash & Linux HPC workflows

ml / informatics

scikit-learn · defect classification
PyTorch · surrogate modeling
OpenCV · SEM image quantification
Materials data pipelines & EDA

geometry / simulation

Implicit surfaces & marching cubes
nTopology-style lattice generation
FEA scripting (Abaqus / ANSYS)
STL / 3MF mesh processing

engineering / ops

Git, GitHub Actions, reproducible runs
LaTeX & scientific reporting
LPBF process telemetry analysis
Data-driven AM parameter design

// computational_projects

Proposed computational work

Pipelines and models planned around metamaterials, shape memory alloys, and additive manufacturing.

⬡

TPMS Lattice Generator

Level-set engine for Schwarz Primitive, Gyroid, and Diamond surfaces. Samples implicit fields on adaptive grids, extracts iso-surfaces via marching cubes, and exports 3D-printable meshes tuned for LPBF minimum-feature-size constraints.

pythonnumpymarching-cubes

🤖

LPBF Defect Classifier

Machine-learning pipeline that maps laser power, scan speed, hatch spacing, and scan strategy to defect likelihood — predicting balling, spattering, and intergranular cracking in NiTi builds before they happen.

scikit-learnclassificationam-telemetry

🌡️

NiTi Phase-Transformation Model

Numerical implementation of shape memory alloy transformation kinetics, linking DSC-measured transformation behavior of LPBF NiTi lattices to stress–strain response for actuator and morphing-structure design.

numerical-methodssma-constitutive

🔬

SEM Microstructure Quantifier

Computer-vision toolkit that segments SEM micrographs of printed NiTi TPMS sections, quantifying porosity, phase distribution, and solidification heterogeneity — replacing manual point-counting with reproducible metrics.

opencvsegmentationimage-analysis

// code_showcase

From equation to print file

The heart of the lattice generator — a gyroid is just three lines of trigonometry until you discretize it, threshold it, and hand it to a printer.

tpms_generator.py — computational-materials
import numpy as np

def gyroid(x, y, z):
    """Implicit gyroid surface — F(x, y, z) = t."""
    return (np.sin(x) * np.cos(y)
          + np.sin(y) * np.cos(z)
          + np.sin(z) * np.cos(x))

def sample_field(level_fn, n=160, threshold=0.35):
    """Sample the level-set field on a regular grid.

    The resulting scalar field feeds marching-cubes
    to produce a watertight, LPBF-ready mesh.
    """
    grid = np.linspace(0, 2 * np.pi, n)
    x, y, z = np.meshgrid(grid, grid, grid)
    return (level_fn(x, y, z) - threshold).astype(np.float32)

# One unit cell, ready for laser powder bed fusion
field = sample_field(gyroid, n=160, threshold=0.35)
print(f"field sampled: {field.shape} · ready for meshing")

// research_with_code

Published work, backed by pipelines

Every study below is supported by reproducible analysis code — from raw process logs to final figures.

2026

Do chopped fibers matter? Process–structure–property relationships and thermal annealing effects in fiber-reinforced TPMS lattices

Investigated the effect of carbon fiber reinforced composite and annealing on mechanical properties of TPMS gyroid cubes.

Progress in Additive Manufacturing · Vol. xxx · pp. xxx–xxx

2023

Microstructural and surface analysis of NiTi TPMS lattice sections fabricated by laser powder bed fusion

Built the image-analysis and statistical pipeline used to quantify Ni evaporation, oxide- and Ti-rich phase formation across Primitive and Gyroid topologies — linking geometry and laser parameters to solid-phase distribution.

Journal of Manufacturing Processes · Vol. 102 · pp. 375–386

2022

Imperfections formation in thin layers of NiTi TPMS lattices fabricated using LPBF

Designed the experimental-parameter sweep and defect-classification workflow that isolated the effects of relative density and scan strategy on balling and intergranular cracking in thin Schwarz TPMS layers.

Materials · 15(22), 7950