nice write up of things that are only obvious if you spend time with AI.
pretty much everything applies to non-agentic AI work, code or not, as well, if you are aiming beyond average quality and conventional design, that is. people who give up somewhat early won't give up much later just because they use AI or teach an AI agent.
but the article is mostly also what people not in the field or tangentially related expect. it's here but that big thing isn't.
I could say I dabbled with woodworking but I really just used a chainsaw to cut down some trees, make slabs and then used drill and screws to construct the cheapest, fastest MVP of a piece of furniture that I used until the shed burned down. But that's not woodworking, not really.
"AI coding agents" is just an autoiterating chat of/with a large coding model, that you still have to iterate over, which is as obvious as an apprentice in a woodworking shop doing a lot--if not all of the work--alone until the meister points out all the mistakes and lets him do it all over again.
> I was soon spending eight hours a day during my winter vacation shepherding about 15 Claude Code projects at once
If you are a "computer person", spending 8h a day on multiple projects is normal, although 15 is, IMO, way too freaking much but I'm ADHD and not really a computer person. While I run dozens of narratives in parallel all the time, I only "shepherd" and iterate over a handful of them in 'flexible' time intervalls.
The reason for the burnout might be, and I can relate due to my ADHD, the following:
> Due to what might poetically be called “preconceived notions” baked into a coding model’s neural network (more technically, statistical semantic associations), it can be difficult to get AI agents to create truly novel things, even if you carefully spell out what you want.
The expectation to create something "truley novel" based on ideas that aren't truly novel (yet, ...what?) is weird enough, but then expecting that an AI coding agent, an apprentice, will make it novel even though the entire thing basically already exists and the novelty makes no sense conceptually until the core elements are separated
> a UFO (instead of a circular checker piece) flying over a field of adjacent squares,
is quite analogues to semi-functional ADHD people who believe they will get at least some of their ideas out if they "work" or dream on all all them. It can work, but you have to separate concerns, which, in case of ADHD people, is becoming functional, meaning don't consume stuff that impede body and brain, do stuff to eliminate bio-physical distractions and to keep hormonal and neural moral high at most times, and only then work, and in the case of AI coding agents it means to separate concepts that are programmatically/mathematically/linguistically intertwined and only then define mechanics and features within or beyond the individual or combined constraints.
> The first 90 percent of an AI coding project comes in fast and amazes you. The last 10 percent involves tediously filling in the details through back-and-forth trial-and-error conversation with the agent. Tasks that require deeper insight or understanding than what the agent can provide still require humans to make the connections and guide it in the right direction.
So then why not at this point switch to the human being primary author and only have the AI do reviews and touch ups? Or are we restricting ourselves to vibe coding only?
> The last 10 percent involves tediously filling in the details through back-and-forth trial-and-error conversation with the agent
It so often sounds like "traditional coding" flows like an orchestra during an opera while vibe and 'agentic' coding flows like a bunch of big bands practicing.
Are they trying to tell the story that "it's the same" or that "it's just not the same"? Is the toolchain changing that much that there is no reason to learn the baseline anymore? So the next ten years of AI development should be left to those who already weild the basic tools? Just like the economy? Is the narrative meant to establish a singularity-driven relationship with young coders, computer scientists and those who use code to entertain, inform and sell via media? While simultaneously pushing the outliers to the edge of the sphere and lock them out via their lack of AI skills and experience with such tools from ever reaching a proper chunk of the mob?
On the one hand, it's a personal decision. Trends and narratives can be convincing. Defactors are rare nowadays while polarization and the status quo are the defacto standard. So on the other hand, it's a depersonalized decision reinforcing the hierarchies (hardware) that dictate which tools (hardware, the cloud) dominate the main stream either way.
> Or are we restricting ourselves to vibe coding only?
> why not at this point switch to the human being primary author
It's the only choice. You are either the primary author of the code or of the learning material. In the former case, the latter is implied and you can't teach if you don't know.
In essence, all this "AI hype" should really only motivate. But these perceptions of "the end of stuff as we know it" and "NOW it's definitely not in my/our hands anymore" that is everywhere weighs heavy. So that the only "residue outcome" really is: making money is the only thing that's left ..., again, reinforcing the hierarchies (hardware) that dictate which tools (hardware, the cloud) dominate the main stream either way--whether you break under the weight or not, whether you shrugg it off or become versed enough to just carry it along--while establishing a singularity-driven relationship of the system with it's constituents.
but the article is mostly also what people not in the field or tangentially related expect. it's here but that big thing isn't.
I could say I dabbled with woodworking but I really just used a chainsaw to cut down some trees, make slabs and then used drill and screws to construct the cheapest, fastest MVP of a piece of furniture that I used until the shed burned down. But that's not woodworking, not really.
"AI coding agents" is just an autoiterating chat of/with a large coding model, that you still have to iterate over, which is as obvious as an apprentice in a woodworking shop doing a lot--if not all of the work--alone until the meister points out all the mistakes and lets him do it all over again.
> I was soon spending eight hours a day during my winter vacation shepherding about 15 Claude Code projects at once
If you are a "computer person", spending 8h a day on multiple projects is normal, although 15 is, IMO, way too freaking much but I'm ADHD and not really a computer person. While I run dozens of narratives in parallel all the time, I only "shepherd" and iterate over a handful of them in 'flexible' time intervalls.
The reason for the burnout might be, and I can relate due to my ADHD, the following:
> Due to what might poetically be called “preconceived notions” baked into a coding model’s neural network (more technically, statistical semantic associations), it can be difficult to get AI agents to create truly novel things, even if you carefully spell out what you want.
The expectation to create something "truley novel" based on ideas that aren't truly novel (yet, ...what?) is weird enough, but then expecting that an AI coding agent, an apprentice, will make it novel even though the entire thing basically already exists and the novelty makes no sense conceptually until the core elements are separated
> a UFO (instead of a circular checker piece) flying over a field of adjacent squares,
is quite analogues to semi-functional ADHD people who believe they will get at least some of their ideas out if they "work" or dream on all all them. It can work, but you have to separate concerns, which, in case of ADHD people, is becoming functional, meaning don't consume stuff that impede body and brain, do stuff to eliminate bio-physical distractions and to keep hormonal and neural moral high at most times, and only then work, and in the case of AI coding agents it means to separate concepts that are programmatically/mathematically/linguistically intertwined and only then define mechanics and features within or beyond the individual or combined constraints.