Blog

When I was a kid, I wanted to write as soon as I found out that books were written by regular people. Hey, I was a regular people! Not too long after, I discovered computers—and developed a second obsession.

I had intended to go into programming as a career. Then, at my first job after college, I accidentally found a job that let me scratch both my writing and programming itches. I’ve been a technical writer ever since.

When large language models (LLMs) arrived, it felt like a technological New Testament, a new era with different rules. The two things I loved and understood best, software and language, had formed a new, more intimate relationship. I’d been using words to write about code; now code could itself read and write. Cool! … Cool?

I spent years documenting AI and ML features at Snowflake and Azure Cognitive Services, close enough to see how the sausage is made, but far enough back to still find it tasty. I understand how LLMs work well enough to have opinions about them. When my knowledge reaches its limit, I know how to find out more.

The Inference exists because I kept having thoughts about AI that didn’t fit in docs. This is where those thoughts go. I’ll try to be technically honest, intellectually curious, and mercifully brief. But no promises on that last one.

The answer to the obvious question: yes, I’m using a large language model (specifically Claude, via Perplexity) to help write this blog. I’ll write more about that soon, when I have more experience with it.