There is a right and wrong way to do Harness Engineering. What makes it right or wrong mostly comes down to whether you're following the lessons of Bitter Lesson Engineering or not. Bitter Lesson Engineering comes from Richard Sutton's "Bitter Lesson" essay, and it means ensuring that you're not trying to outsmart your own AI. It means not trying to micromanage how your AI does things, but rather specifying what you want done. Plainly stated: Bad Harness Engineering is a whole bunch of prescriptive instructions on exactly how to do things. First copy this file, then load this, then do this, then do that. Etc. Good Harness Engineering is about providing tons of context about who you are, what you're about, what you're working on, what you're trying to accomplish, and what good (and bad) look like to you. I'm an engineer focused on personal productivity, I like simple designs with lots of whitespace and great typography, here are my previous projects, here are some tools you can use, etc. I have a Bitter Lesson Engineering skill that I run often against my entire harness to audit for this. Bad Harness Engineering is bad because the smarter AI gets the more antiquated your instructions will become. And at some point (maybe even now?) they'll make your AI stupider instead of smarter. Good Harness Engineering is good because no matter how smart an AI becomes it will still be better at getting you great results if it understands who you are and what you like. Also, the better AI gets, the more important Bitter Lesson Engineering becomes. Basically, both your prompts and your harness should be about who you are and what you're trying to accomplish, and not specifically how to get there. That's what the AI is there to figure out. Give it the best possible picture of you, your ideal outcome, and the best tools you can, and give it room to work. Notes There are, of course, lots of other ways to do good and bad Harness Engineering. This is just, in my opinion, the biggest distinction and pitfall to be aware of. In the early (2023-2024) days of Prompt Engineering it was ok to tell AI exactly how to do things, and it actually helped. Who knows exactly when the inversion happened, but it was probably somewhere in 2025. I don't mean to imply in the Good Harness Engineering section that it should be super terse. My harness has vast amounts of context, but it's about what I want, not how to achieve it. This principle applies to prompt, context, and harness engineering, and both personal and Enterprise AI. The BLE Skill is available in PAI . Citation: Richard Sutton, "The Bitter Lesson" , March 13, 2019. Related: "Bitter Lesson Engineering" — my own take on applying Sutton's lesson to AI system design. AIL Level 1 (0%): Daniel wrote this post entirely. I (Kai, Daniel's assistant) helped with formatting and publishing. Learn more about AIL