Mike ONeal

I started programming in the 1980s. Not with JavaScript or Python — with assembly language and C. Memory management. Pointers. Registers. The kind of programming where a single off-by-one error doesn't throw an exception — it corrupts memory and crashes the machine.

From there, I watched every layer of abstraction get built. I didn't just watch — I built on each one as it arrived. Desktop applications when GUIs were new. Web applications when HTTP was still a novelty. Mobile when the iPhone SDK dropped. Cloud infrastructure when AWS was just S3 and EC2. Machine learning when you still had to write your own gradient descent.

Each era taught something the next one needed. Systems programming taught me how computers actually work — not the abstraction, the metal. Web taught me distributed systems and the reality of network latency. Mobile taught me constraint-driven design. Cloud taught me infrastructure as code and the economics of scale. ML taught me statistical thinking and the difference between a demo and a production system.

This is why I can build AI systems that work in production — because I understand everything underneath. When an LLM hallucinates, I know why. When a pipeline fails at scale, I know where to look. When someone says “it works on my machine,” I know exactly what that means and exactly how to fix it.