I've spent a lot of time on context engineering my AI agents. This often leads me to generalizing how good AI agents are out of the box.

I'm reminded when I take the exact same prompt and use it in any other place and the results are really bad. Then it makes sense why people are so skeptical of AI