Post

David Heath
@davidheath.org

All of this makes a lot of sense. I’m at the very start of my journey in this and will need to guide others. Thanks for sharing.

David Heath
@davidheath.org

bsky.app/profile/davi...

David Heath@davidheath.org

just gave github.com/DrCatHicks/l... a spin on a hobby project I was working on this afternoon. It challenged me to interrogate the design in ways that I hadn't bothered to do. Very impressed at the questions! I can see how this would help people stay familiar with the code they work on.

github.comGitHub - DrCatHicks/learning-opportunities: A Claude or Codex skill for deliberate skill development during AI-assisted codingA Claude or Codex skill for deliberate skill development during AI-assisted coding - DrCatHicks/learning-opportunities
iamwil
@interjectedfuture.com

1. Interrogation/interview-based design and specification 2. Make the decisions that matter and leave the rest to the agent. 3. Use agent to generate explainers, quizzes, and simulations. 4. Use verifiers and guardrails in the coding loop.

iamwil
@interjectedfuture.com

Interview-based design and spec: When I tell the agent what I want to build or extend, there's naturally ambiguity. But the agent knows what is ambiguous to it, so it can ask me. Closing that specification gap is how you get better results from your agents.

iamwil
@interjectedfuture.com

Make the decisions that matter: Not every decision matters in a product. Agents are fast at writing code and pretty good at some of the nitty gritty details, so I let it make those decisions. I make decisions on architecture, approach, and stack that will help differentiate.

iamwil
@interjectedfuture.com

Generating explainers, quizzes, and simulations: Asking for explainers helps quite a bit, though I have a custom skill for it. Agents aren't great at it out of the box. I also have it quiz me to help retain the concepts I learned from explainers.