AI编码危机：团队丧失架构认知与维护困境

The Problem is not the AI Code, but Nobody Knows Anything Anymore

Last updatedUpdated: Sep 28, 2026 by Simon Späti · CreatedCreated: Sep 26, 2026 · 4 min read recently updated Recent changes Sep 28today published · 882 words

If we think Is writing code dead, and AI is generating all codebases, I still think the bigger problem is people or full teams not knowing anything anymore about the system architecture or the intent behind why certain choices have been made.

A comment on a discussion I had:

I think AI writes probably average code (depending on the task and size). So if your code base was below average AI can easily improve it up to average. At least that’s what I’ve observed here.

To me, the problem is not the AI code, but that nobody knows anything, and everyone just asks Claude. You end up with no plan whatsoever.

The Current State in Fast Moving Startups

This tweet summarizes the current state at fast moving startups well, or larger companies or where middle management is pushing AI hard:

I am done with this shit. It is over. The state of engineering right now is horrible. It has been half a month since I started a new role at a big company. Nobody knows anything here. The specs, code, tests, PRDs, tickets, resolution of those tickets, reports, etc., everything is made by Claude Code.

Nobody on my team likes this. They are being forced to ship as much as they can. I have heard multiple times from higher management that pushing code is not a bottleneck, so why are we slow? People are working 12 to 13 hours a day just to press enter. Nobody is reading anything. Humans in corporate are doing nothing on their own.

Everyone, literally everyone, from an L1 to an L7 engineer here is doing the same thing. Talk to Claude. There is no sense of victory. Nobody is resolving bugs. In reality, nobody is thinking anymore. Everything is done by LLMs.

It is so soul-sucking. I would not mind it, to be honest, if we were at least given the time to check out the code and see what is going where. But no, the goal is to just ship. No matter what happens. Voxium

Data Engineering is Different?

Hoyt Emerson mention that data engineering is different:

I think Data people are different. We’ve had to know everything about the product/business from day 1. AI just removes friction for us now. Tweet

I think data people who grew up pre-AI had to know everything (or a lot, or involve domain experts) to figure it out, indeed. But AI makes this obsolete, or seemingly obsolete.

That’s why people starting today, or me as well, if I start today prompting away in a new field, all of a sudden, that knowledge is missing.

A Product Manager Could now Build Anything He Wants

Good point by Sean Behan:

I’ve always admired product people who can’t code but can manage a team to get the software they want. Knowing what you want has always been the hardest part.

One could say a good product manager could now build anything they want and find a market, make it look good, etc. But then again, if you can’t code, you will essentially build a very bad foundation for a product that’s very hard to maintain (although AI is getting better at that too, especially when you iterate often, but still, if you choose the wrong language or the wrong mental model, you have the wrong start from the get-go).

It still helps to know the fundamentals, either way: for programming and designing a product, and for a good PM who knows what is needed but also understands system and architecture design.

The Final Boss is Still Maintenance

Thinking in systems, or architectures, or having intent and design- all of them help to be a better software engineer. Nowadays, Writing code by hand might be dead, but it certainly helps, and Having Taste (with AI) is more important than ever.

But the final boss is, and always will be, maintainability. The easier it is to generate a quick pipeline, app, or BI dashboard, the more you have to maintain. And if nobody knows a thing, that can get really hard.

AI Can’t Drive Itself

Yes, the AI can’t prompt itself, right? Why do we even need humans? To me, it’s a clear sign that humans are still needed to direct and orchestrate it. That’s also why intent, taste, design, and architecture are all killer features in today’s world.

But once these are absent, or even worse, fundamental, get lost, it’s really dangerous. I read today that this is a self-inflicted problem, and if we still hired juniors, then the problem wouldn’t be happening. But yeah, it’s not as easy.

Further Reads

• Kris Jenkins talking about the worry of middle management, not the vibe coding at Danger of AI or LLMs
• If Consumed by a Human, Should Be Written by Human
• What I Learned Writing with AI

Origin: the primagen video and Limitations of LLMs and AI

References: What I Learned Writing with AI

Data Engineering & Second Brain Newsletter

This note is part of my Second Brain, a 650+ public notes on data engineering, life, and more. I send a digest of new notes, blog posts, and book chapters, and general updates by email. Subscribe below if you'd like to get the next one.

Interactive Graph