10 Clean Architecture questions I wish I had answers to sooner 👇 youtu.be/10b8sJomdCI

With AIs “distance” is important. One AI that writes tests, and another that makes them pass, sets the two at odds with each other and prevents either from cheating. This is very much like the separation of powers in a democratic constitution. Another form of distance is

Unpopular truth: We trust humans because we can hold them accountable. Jobs. Contracts. Reputation. So they’re motivated to deliver quality and avoid cutting corners. LLMs have none of that. That’s a fundamental difference 🤷‍♂️

When we start testing apps manually again, as many of you have suggested 👇 AI has successfully outsourced QA back to us developers 🤷‍♂️ x.com/plainionist/st…

I've seen a lot of posts complaining that AI is non-deterministic. This is true, but my experience is that AIs can be constrained to be very nearly deterministic. Some might say "very nearly" is not good enough. My response is that I believe I can crank up the constraints to

A lot of the lessons my son and I have learned about AI over the last few months are described in videos on cleancoders.com. Look for Agentic Discipline.

Hot take: Software engineers will soon write more specs than they read. Specs for acceptance criteria. Specs for architecture. Specs for everything. AI handles the implementation.

If you think software engineers will be replaced by LLMs, you’re not a software engineer. Sorry 🤷‍♂️

Unpopular opinion: LLMs don’t increase technical debt - they help reduce it. I can now do large-scale refactorings in minutes, alone - work that used to take a team days 🤷‍♂️

Most countries require a driver's license to operate a car. Do you think developers should need a "vibe-coding license" to use LLMs? 🤔

Why aren’t you using test-driven development (TDD), knowing it results in fewer bugs and better design?🤔

When you vibe-code, do you still take ownership of the software AI writes? 🤔

Some say: "AI fails at complex tasks." I’d say that’s your job as a software engineer: Break the problem down into a sequence of small, simple steps so AI can execute them autonomously 👍😉

Refactorings are important and should be integral part of software engineering just like writing automated tests. But of course each refactoring must have a purpose, it must add some (business) value - like everything else.

Hot take: Delegating tasks to AI is not the game changer. Having a personal tutor in your pocket is. Ask anything. Debate ideas. Go deep - anytime. That’s how you grow faster. That’s the real game changer.

How do you run developer interviews in the age of AI? You’re not still asking people to code on a whiteboard ... right? 🤔

Unpopular opinion: Your tech stack barely matters anymore. Frameworks and libraries are just details - and LLMs can handle them more effectively than you ever could.

What was the worst code you’ve ever seen a developer commit? 🤔

Humans are not supposed to be "compilers" anyway 🤷🏼‍♂️😉

Hot take: Vibe coding is like fast food: It’s convenient and fine occasionally. But unhealthy in the long term. Change my mind 🤷‍♂️