Blog & Articles

Blog & Articles

Subscribe to learn more about building with AI and get the newest experiments and lessons from shipping in public.

Latest writing

Recent Posts

The 3 ways to grade LLM outputs automatically (and when each one fails)

Does your prompt actualy work?

Testing a prompt on one example is like shipping code with one unit test.

I stopped guessing whether my prompting was any good and started scoring it

Prompt evaluation intro.