I work at the intersection of machine learning and materials science.
My current focus is on using data-driven approaches to understand how materials behave - especially in problems related to degradation, performance, and electrochemical systems.

Most of my work so far has been on building predictive models from experimental and engineering data. Recently, I’ve been shifting toward questions that are closer to computational materials science - how we can use ML alongside atomistic methods to study processes like catalysis and charge transport.

• Using machine learning to learn structure–property relationships
• Electrochemical systems (corrosion, catalysis, energy materials)
• Moving from experimental datasets to computational/atomistic modeling
• Building simple, reliable models that actually explain something

• Predicting fatigue strength of alloy steels using ML
• Analyzing corrosion behavior from experimental data
• Studying how processing conditions affect material performance

These projects are small, but they all follow the same idea:
start with real data to build a model → extract insight, not just accuracy

Right now, I’m working on strengthening my foundation in computational materials science - especially atomistic modeling and its integration with machine learning.

I’m particularly interested in problems like:

• electrocatalysis
• reaction mechanisms at surfaces
• data-driven approaches to energy materials

I’m still early in this transition from applied ML towards computational materials.
This GitHub is a record of that progression - from data-driven engineering problems toward more physics-based and atomistic approaches.