← Back to Portfolio

Writing & Notes

Thinking out loud about AI research, federated learning, LLMs, and the process of becoming a researcher.

Backpropagation: The Algorithm That Teaches Neural Networks

Backpropagation is the engine behind all of deep learning. It's not magic — it's calculus, applied carefully. Here's the intuition, the math, and why it actually works.

Read post →

Overfitting & Regularization: When Your Model Knows Too Much

A model that memorizes is not a model that understands. We explore overfitting — what it is, why it happens, and the elegant mathematical tricks we use to fight it.

Read post →