About Me

I am a PhD in Computer Science with more than a decade of experience in machine learning and applied AI systems. My work sits at the intersection of mathematical modeling, statistical learning theory, and modern large-scale neural architectures.

I currently work as an Innovation Manager (and previously, a Senior Data Scientist) at Experian, where I design and deploy machine learning systems under real-world constraints. My interests are centered on the theoretical structure of learning systems and the behavior of models in high-dimensional regimes.

Background

My doctoral training provided a rigorous foundation in probabilistic modeling and inference. Over time, my focus has expanded toward broader questions in machine learning theory and modern AI:

  • Generalization in overparameterized models
  • Optimization dynamics in deep networks
  • Implicit regularization and inductive bias
  • High-dimensional statistics
  • Representation learning
  • Transformer architectures and foundation models
  • Scaling behavior of neural systems

I am particularly interested in understanding why large neural models work as well as they do — and where current theoretical explanations remain incomplete.

Professional Work

Over the past ten years, I have built production ML systems across structured and textual domains using the scientific Python ecosystem.

My applied work includes:

  • Designing large-scale predictive systems
  • Evaluating models beyond standard validation pipelines
  • Integrating deep learning models into operational workflows
  • Stress-testing systems under distributional variation
  • Bridging theoretical assumptions with industrial constraints

Modern AI systems often succeed empirically before being theoretically understood. Much of my interest lies in studying that gap.

On Modern AI

The rise of deep neural networks, transformers, and large language models has shifted the center of gravity of machine learning toward large-scale representation learning and emergent behavior in overparameterized systems.

Questions that motivate my work include:

  • What governs generalization in highly overparameterized regimes?
  • How does optimization shape learned representations?
  • What structural properties emerge in large transformer models?
  • How do scaling laws relate to statistical efficiency?
  • Where do theoretical guarantees fail in practice?

These are fundamentally mathematical questions, even when the systems are empirical.

This Blog

This blog collects technical notes on machine learning theory and modern AI systems.

The emphasis is on:

  • Mathematical structure of learning algorithms
  • Theory-informed analysis of neural models
  • Limits of current theoretical frameworks
  • Structural properties of transformers and LLMs
  • High-dimensional phenomena in modern ML

The goal is depth rather than breadth.

For discussion or collaboration, feel free to reach out via GitHub or LinkedIn.