Stephen Kerr

Computational Materials Science

PhD candidate in computational materials science at Queen's University. Studying vacancy–solute coupling in nuclear alloys (Ni–Cr–Al, Zr) and CO₂ catalysis using DFT, machine-learning interatomic potentials (MACE, ACE), and kinetic modelling (k-ART).

Now: MACE-driven k-ART trajectories in Ni–Cr–Al · CO₂ binding on stepped Cu · Zr–Nb defect benchmarking

Teaching

I teach and support undergraduate courses, labs, and tutorials with a focus on clarity, reproducible computation, and constructive feedback. Students progress from analytical solutions to Python scripts that test assumptions, check residuals, and quantify numerical error. My teaching draws directly on my research — students in computational courses work with the same tools and verification practices I use in atomistic simulation, building habits that transfer to research and industry.

Teaching Philosophy

My approach centres on bridging theory and practice. I set clear expectations and transparent rubrics early, then work alongside students to build confidence with both analytical and computational methods. I emphasize reproducibility—version control, documentation, and verification habits—because these are the same skills that underpin reliable research. Formative assessments and peer explanation keep feedback constructive and timely.

I am particularly interested in responsible AI integration in STEM coursework. Students now have access to LLM-based tools that can generate code, solve textbook problems, and draft reports. Rather than prohibiting these tools, I design assignments that use them deliberately—asking students to verify AI-generated solutions against physical constraints, critique model outputs, and document when and why a tool fails. This builds the critical evaluation skills that matter in research and professional practice while acknowledging the reality of modern computational workflows.

Teaching Interests

Numerical methods in Python Data fitting & uncertainty Active tutorials Problem-based learning Rubric-based feedback Responsible AI in STEM education Inclusive teaching Computational materials labs

Teaching Experience

Queen's University Current

2024–present

MECH 479 Nanomaterials

Instructor + TA, 2024–2026, 20–40 students. Interactive lectures on materials simulation, molecular dynamics, and computational nanoscience. Designed cross-platform tutorials emphasising reproducible workflows, version control, and verification of numerical results — including a new computational module on MD with LAMMPS and post-processing in Python. Course content connects directly to my research in atomistic simulation of defect behaviour in alloys, so students work with the same DFT and MD tools used in active research projects.

MECH 272 Materials Science

Lab TA, Winter 2025–2026. Undergraduate materials laboratory supporting hands-on characterization and testing methods.

CHEM 112 General Chemistry II

Lab TA, Winter 2025–2026. Laboratory instruction for first-year engineering and science students.

Ontario Tech University Previous

2021–2024

Winter 2023 — Full TA Load

Approximately 1,600 reports and quizzes marked in one term across:

  • CHEM 1800U Chemistry for Engineers (9 lab sections, ~20 students each)
  • CHEM 1020U Chemistry II (4 lab sections)
  • CHEM 1010U Chemistry I (2 labs + 4 tutorials, ~20–45 students each)
  • Science Café (weekly drop-in support, 5–15 students per session)

Additional terms included coordination of multiple CHEM 1010U and 1020U lab sections, development of comment banks for efficient feedback, and mentoring of undergraduate peer tutors.