What a brilliant piece. The way you connected Kahneman's System 1/2 framework to AI-aided development — and then made the counterintuitive case that faster tools demand slower thinking — is both sharp and timely.
I deeply resonate with your article. I've been developing the same core idea independently for over 25 years — that slowing down is the real accelerator, especially now in the age of AI-aided programming.
I wrote a series called "Don't Go Too Fast" where I built practical tools around this exact insight: how to find the right problem before coding (IWD), how to design complete service chains (ROD), how to treat requirements as tests (TFD), and how to maintain System 2 thinking under pressure (DGTF).
Your observation that "AI can elegantly implement thousands of lines of code for the wrong requirement" is what I call "the happy journey to the wrong destination" — and it's the central fear that started my entire framework.
I'd love to discuss this topic with you. Here are my works:
Amusing how we're now rehashing software engineering basics, only with AI as the target. Make a plan! Clarify specifications! Write requirements! Figure out the drawbacks! This has always been the foundation of good software engineering.
Hi James,
What a brilliant piece. The way you connected Kahneman's System 1/2 framework to AI-aided development — and then made the counterintuitive case that faster tools demand slower thinking — is both sharp and timely.
I deeply resonate with your article. I've been developing the same core idea independently for over 25 years — that slowing down is the real accelerator, especially now in the age of AI-aided programming.
I wrote a series called "Don't Go Too Fast" where I built practical tools around this exact insight: how to find the right problem before coding (IWD), how to design complete service chains (ROD), how to treat requirements as tests (TFD), and how to maintain System 2 thinking under pressure (DGTF).
Your observation that "AI can elegantly implement thousands of lines of code for the wrong requirement" is what I call "the happy journey to the wrong destination" — and it's the central fear that started my entire framework.
I'd love to discuss this topic with you. Here are my works:
GitHub: https://github.com/blueist/dont-go-too-fast
Medium: https://medium.com/@stillblueist
It's rare to meet someone who arrived at the same conclusion from a different path. Looking forward to connecting.
Best,
Bakmeon Kim
Hey! Thank you for the kind words and also for sharing your own work.
“The happy journey to the wrong destination” is a better phrase for it than anything I came up with. 🙂
Amusing how we're now rehashing software engineering basics, only with AI as the target. Make a plan! Clarify specifications! Write requirements! Figure out the drawbacks! This has always been the foundation of good software engineering.