An AI engineer's notes on technology, cognition, and how systems—software and human—really work.
Every laser ever built works because photons can clone themselves. A deep dive into how we get from a light bulb to burning drones out of the sky, starting with Einstein's 1917 prediction that nobody knew what to do with for 43 years.
The neuroscience behind phone addiction: how Skinner's pigeons, dopamine prediction errors, and the wanting-vs-liking distinction explain why you can't stop scrolling.
How Codex, Aider, RooCode, and Cursor apply AI-generated changes to files—comparing patch formats, fuzzy matching, and Cursor's specialized apply model.
America's 'wild' horses are actually feral descendants of Spanish imports. A ChatGPT-assisted deep dive into 55 million years of horse evolution and extinction.
Jeff Bezos on two-way doors, why senior leaders should speak last, the danger of proxies, and how 6-page memos beat PowerPoint for truth-seeking.
From Aristotle to Elon Musk: how breaking problems down to fundamental truths and reasoning up from there leads to breakthrough innovations.
Why uv is now the best tool for Python environments, plus the pyenv + venv workflow that keeps each project isolated with its own Python version.
A practical guide to NDCG, MAP, Precision, Recall, and MRR—the metrics used to evaluate search systems and neural retrieval models, with worked examples.
How Tesla, Netflix, TikTok, and even a local butcher use the virtuous circle of AI—where users generate data that improves the product, creating an unstoppable competitive advantage.