May 2024 – Sep 2024 · 4 mos · Pittsburgh, Pennsylvania, United States · Hybrid

• Optimize nano-structured thin films for optical filters and TPV systems, achieving high reflectance at long wavelengths.

Aug 2020 – Oct 2025 · 5 yrs 2 mos · Pittsburgh, Pennsylvania, United States · On-site

• ◦ Machine learning based Monte Carlo simulation of ferrites
• ∗ Conducted DFT calculations on 500+ different ferrites and developed code to organize all the calculation data into well-structured SQLite format files.
• ∗ Represented structures of ferrites based on bond analysis as 1D vectors for predictive modeling.
• ∗ Fine-tuned hyperparameters using random search and evaluated the performance of multiple candidate machine learning models (linear regression, neural network, gradient boosting machine and support vector machine) using test RMSE of four-fold cross validation.
• ∗ Accelerated Monte Carlo simulations by embedding trained machine learning models.
• ∗ Predicted temperature dependent cation distribution and magnetization for a variety of ferrites, contributing to the design of complex ferrites with better magnetic and optical properties.

Aug 2018 – May 2020 · 1 yr 9 mos · Pittsburgh, Pennsylvania, United States · On-site

• ◦ Thermodynamic modeling of LiFePO4 − FePO4 system
• ∗ Performed DFT calculations on LiFePO4 and FePO4 for ground state energies using Quantum Espresso and designed different sublattice models (CALPHAD modeling method) to describe LiFePO4 − FePO4 system and
• generated phase diagram via software Themo-Calc, which fit well with experiments.
• ◦ High entropy alloys as a discovery platform for electrocatalysis
• ∗ Performed DFT calculations to calculate the adsorption energies of ∗OOH and ∗OH on the HEAs (Ir, Pd, Pt, Rh, Ru) surfaces and conducted feature engineering on the atomic environment of the adsorbate site to construct
• dataset.
• ∗ Derived linear relation of atomic environment - adsorption energy and predicted adsorption energies of new adsorbate sites, found good agreement between DFT calculated and predicted values.

Sep 2017 – May 2018 · 8 mos · Hangzhou, Zhejiang, China · On-site

• Synthesized high-purity nano-silicon to be used as the negative electrode material for Li-ion batteries. Employed controlled magnesiothermic reaction to ensure uniformity and quality control.
• Assembled Li-ion batteries with nano-silicon as the negative electrode.
• Utilized X-ray Diffraction (XRD) techniques to characterize the crystalline structure of the synthesized nano-silicon and Scanning Electron Microscopy (SEM) to visualize and assess the morphology of nano-silicon particles.
• Evaluated the assembled Li-ion batteries for electrical performance, including capacity, rate capability, and cycle stability. The nano-silicon-C composite electrode exhibited (a high initial capacity of 700 mAh/g at 100 mA/g and maintained 82.5 of its capacity after 100 cycles at 1C rate).