About

I'm Aaron Gayle. I came up through digital marketing and ad tech — attribution modeling, A/B testing, audience segmentation, campaign analytics, the daily habit of tying everything back to ROI. Somewhere along the way I noticed the job had quietly become more about data and automation than the title admitted. Polynomial World is where I document the pivot into AI, data, and engineering — in public, as it happens.

I'm a practitioner, not an academic, and I don't have a CS degree — I'm doing this self-directed, learning by building rather than collecting credentials. As my projects are far enough along to share, they'll show up here. The writing is first-person and anti-hype: I care as much about where AI tools break as about what they can do, and if a piece doesn't have a point of view, it doesn't belong here.

What you'll find

• Career — the honest mechanics of changing fields into AI/data without starting over.
• AI & LLMs — building real things with models, and where they fail.
• Data & ML — learning the craft in public, from first Kaggle notebook onward.
• Engineering — making a demo trustworthy: evals, monitoring, shipping.

I'm based in Tokyo (though I'm not Japanese-bilingual, so that's context, not the story). If you're making a similar move, the articles are written for you as much as for me.

Elsewhere

Polynomial World is also the umbrella for my side projects — see the list on the home page. My portfolio lives at aarongayle.com.

Questions or feedback? Get in touch.