Notes on building AI systems, tools, and things I learn along the way.
August 24, 2026
"Follow your passion" is the default career advice. I used to buy it too: figure out what you love, then go build a career around it.
August 17, 2026
Every LLM you've used runs on this idea. Attention is the mechanism that lets a model figure out which parts of the input actually matter for the prediction...
August 10, 2026
A neural network doesn't know anything when you create it. Every weight starts random. Everything it "learns" comes from a loop: make a prediction, measure h...
August 3, 2026
More features isn't always better. A dataset with hundreds of columns often has a much smaller "true" structure hiding inside it, plus a bunch of noise that...
July 27, 2026
No labels, no "right answer" to check against. Unsupervised learning is about finding structure in data when nobody's told you what the structure is supposed...
July 20, 2026
Training a model is the easy part. The stuff that actually separates a working model from a broken one: picking the right metric, tuning it properly, handlin...
July 13, 2026
Here's SHA, SimHash, and Diffie-Hellman, and where each one actually gets used.
July 6, 2026
Not every algorithm fits into "sort this" or "find the shortest path." Some solve completely different problems: breaking a signal into its parts, running wo...
June 29, 2026
Type a word into a search bar and get results in under a second, out of billions of documents. That's not magic. It's a handful of data structures doing very...
June 22, 2026
You don't need to memorize every algorithm out there. You need to know which tool fits which problem, and be able to explain why.