teaching

Practical machine-learning material developed while assisting Dr Sathiskumar Ponnusami at Queen Mary University of London.

I authored and maintain these tutorials as reusable teaching assets for mechanics-focused AI and scientific machine learning audiences.

teaching footprint

The portfolio below focuses on material quality, reproducibility, and clear learning outcomes, rather than polishing-only presentation.

Course-facing focus Runnability-first workflows that combine short conceptual explanations, notebook execution, and practical follow-up examples for self-study or classroom use.
Ownership and upkeep All published tutorial content is authored and maintained by me; each page includes authorship and maintenance notes so reviewers can verify contribution history quickly.
Role fit with your profile The current material sits at the intersection of phase-field mechanics, ML modelling, and explainable physics-informed workflows in engineering contexts.

These are the current tutorial hubs currently in active maintenance.

PyTorch Tutorials Hands-on model-training workflow from loops and autograd fundamentals to applied physics-informed examples.
Machine Learning Fundamentals Regression, classification, model diagnostics, and generalisation habits for a clean starting point.
Scientific ML Mechanics-aware ML applications that focus on interpretation and model behavior under constrained engineering data.

Student-friendly and reviewer-friendly assets you can use directly in an interview discussion.

Core tutorial stack Five structured tutorials that connect training mechanics to PINN workflows, each with runnable assets.
Notebook-first delivery Each resource is built to open as code-first learning with clear intermediate outputs, making it easy to validate understanding step-by-step.
Publicly cited references The material references standard engineering and ML primers for continuity, with explicit credit notes when external explanations are adapted.

For recruiters Use this page as the teaching evidence node: ownership, context, and practical outputs are explicitly stated.
For collaborators Bookmark the hub and choose one path based on audience: PyTorch fundamentals, ML foundations, or scientific ML.
For your own study tracker The material is designed to map to both foundational ML concepts and applied mechanics examples in a single progression.