Why?

I'm Ed, a chemical engineer turned AI engineer who can't stop asking why.

My background is in process engineering: oil and gas, optimization, systems that flow. I spent years thinking about how things transform, inputs to outputs, energy balances, feedback loops. That lens never left me. When I moved into machine learning, I brought the same obsession: not just what works, but why it works.

This is Grokking Machines, a place where I work through the fundamentals of how machines learn. Not tutorials. Not quick wins. Just the slow, honest process of understanding.

"Grokking" means to understand something so deeply it becomes part of you. That's what I'm after. When I learn back propagation, I want to know why gradients flow backward. When I see a loss function, I want to know what it's actually measuring and why that matters. I keep pulling at threads until the whole thing makes sense.

Sometimes I'll start from first principles and build up. Sometimes I'll build something first, then tear it apart asking why until I hit bedrock. Either way, the goal is the same: real understanding, not surface fluency.

There's no curriculum, no roadmap. Just my notes on the slow work of actually understanding. A learning memoir, if you will. If you want the why too, you're welcome here.