Mathematics · AI Systems · Cross-disciplinary Synthesis

Working at the intersection of theoretical mathematics, system-aware machine learning, and structural reasoning.
My current focus is on how mathematical structure, representation, and constraints shape learning dynamics in modern AI systems.

I treat learning as a cumulative, publicly inspectable process rather than a collection of isolated outcomes.

• Mathematical Foundations
  Studying optimization, probability, and numerical methods with an emphasis on how formal structure translates into computable systems.

• Model Architecture & Learning Dynamics
  Investigating Transformer-based models through minimal, transparent implementations to understand training behavior, representation formation, and scaling effects.

• Cross-disciplinary Structure Mapping
  Exploring how ideas from mathematics, computer science, and the arts share latent structural patterns, documented through long-form learning records.

• learning-journal
  A continuously updated, public log of technical reasoning, conceptual refinement, and long-horizon learning trajectories.

• Math-381
  Mathematical explorations focused on numerical analysis, stochastic processes, and optimization-oriented thinking.

• nanoGPT
  A systematic study of minimal Transformer implementations to analyze training mechanics and architectural decisions.

• MedSync
  An applied project examining how AI components integrate into real-world, constraint-heavy system design contexts.

• Mathematics & Modeling
  Linear Algebra · Probability · Numerical Methods · Optimization

• AI / Machine Learning
  Neural Architectures · Training Dynamics · Representation Learning

• Programming & Systems
  Python · C++ · System-oriented Engineering Practices

• Email: (add a dedicated academic/technical email here)
• GitHub: https://github.com/WZatU

Theory-informed engineering. Structure-first problem solving.