Context is the scarce resource — and your AI still can't see your website

The useful part

This one is about the thing that actually determines output quality — and a feature we shipped because of it. That property is the whole reason we can test it, cache it, and reason about it. If you've built anything real on top of an LLM, you already know this in your bones: the demo works, the eval is flaky, and the difference between "ship it" and "delete it" often comes down to something you can't quite point at.

How it works

• How much of the actual situation the model can see at the moment it answers.
• Run A gives you a plausible, generic answer that assumes a framework you're not using and an API shape you don't have.
• Context is the dominant variable The honest answer, after a couple of years of building on this stuff, is context.

Example or evidence

Details worth keeping

The clearest way I know to feel this: take the same task and run it twice. # Run A — no context "Add pagination to the product list." # Run B — same task, repo in context "Add pagination to the product list." #...with the actual ProductList component, the API client, # and the existing useProducts hook in the window. Same model, same temperature, same prompt string.