Kasim made a distinction on a call that changed how I think about AI.
Prompt engineering is top down. You write a clever instruction and you get the better version of the average.
Context engineering is the opposite. You give the AI everything it could possibly need, from as far back as you can go.
Some people on that call have been in business for 20, 30, 40 years.
All of that experience can be loaded into the tool. Then, in his words, it “fires and it cooks.”
He also said something that stuck with me: founders shouldn’t be the ones learning every AI trick. That’s not the needle mover.
The needle mover is knowing how to deploy it, how to use the people who use it, and how to manage to it.
👇 What’s the one piece of your business knowledge that your AI still doesn’t have?