I wish people would describe in more detail the tasks they use LLMs to code. My experience is that simple components in an existing architecture are fine, but anything requiring architectural considerations quickly becomes a mess. On my projects (e.g. a ui framework), running multiple agents in parallel would just increase the speed at which it can stuff up the project.
i've been running claude in what the blog calls phase 0 for the last 6-7 months. i'm perfectly happy with it, my development velocity has increased while i still have a good grasp of the entire app, and i've actually been making decent progress with web development for a personal project, which is something i've bounced off several times in the past. also i do not get stuck as often on stuff like "how do i get django to statically serve up a js bundle with relative imports" which is more about knowing specific APIs of specific frameworks than any feature of my code or architecture.
i would not want to go down the "take myself out of the loop" path because yes, i do have to micromanage the claude session, often course-correcting every commit and then doing large scale refactoring every so often. but i'm perfectly happy doing that - i see claude as more of a tool than a coder i can hand work off to.
I'm currently using it to do a large migration from one Relay environment to another, but this is possible because
1. We've done it by hand for another route already, which the LLM uses as reference
2. Theres a strong validation setup/harness I've setup for it with storybooks, and component tests
3. It's a _mostly_ mechanical transform. Not entirely, as the two environments/APIs are not 1:1, but it's close enough
But! I and my team are still reviewing everything shrug it is "faster" because I get to have this running while I'm in meetings planning other more interesting projects
And this isn't really that many agents in parallel. Yeah, plenty of fan-out subagents, but that IMO doesn't count/isn't really the same as what others are talking about
In the last week we have done a complete analytics dashboard overhaul with Fable/Opus. The baseline was really bad, for we have no front-end engineers, so we largely felt comfortable not reading anything but the auth code (where we did find one subtle edge case handled incorrectly).
The pipelines and data serving design was all human since it did have to deal with some data scale but the javascript/api layer was all slop, and it seems fine and good.
If you have a really high quality piece of code that needs to meet a high bar of quality/reliability, then I think the risk of letting the AI loose on it is very high and I wouldn't do it. If you have a pile of code you already know is a pile of garbage despite being human written, well, it can't get much worse :)
I also built an agent orchestration meta harness that runs on k8s and uses the k8s agents sandbox for running codex/claude code in the cloud. This was almost entirely just handed over to Fable and I have not asked a single architectural detail. The quality of this product is mediocre, but the fact that it largely works after I went through a few iterations of clicking around is impressive. I would have preferred to buy something off the shelf, but nothing even really came close (though maybe now I would have forked Omnigent)
I don't know if I’m overly critical but there’s gotta be a middle ground between totally AI pilled people that otherwise have no talents, and control freak veteran developers who cant let go
My current process is also using Github projects in a normal scrum style way, with many tickets written or fleshed out and state managed by the LLM, and it doubling as the memory system
Completely leapfrogging all these other open and closed source concoctions and being more effective
But its effective enough that I don’t need OP’s final form state of still approving everything
Auto-mode is fine. Worktrees are built into Claude Code now. I just tell it to classify tickets as sequential or parallel possible and spawn subagents to tackle all of the tickets in the todo list
They all get their own context window its pretty perfect now
in the meantime I work in a couple tabs of Claude Design for different flows of any client side app. My philosophy has been that devs could pick up graphic and UI/UX design easily, its just still a full time job to make variations of layouts and portray their states.
UI/UX is not a full time job anymore.
And I use Claude chat to flesh out aspects of the overall idea
I think you may be overcomplicating your workflow in the concluding state.
Overall I agree that planning and intention is now most of the time, before a 10 subagent precision strike is initiated
I just do turn based development with Cline. I design my UIs in the browser first then let something like Claude wire it up, correcting it as I go. Way faster than before, easy to correct mistakes, doesn't require self-sacrifice or submission.
I shudder when I hear about some people's (wildly overcomplicated) setups. I get the allure but there's something nice about pair programming with an LLM in a singular chat.
Interestingly, despite it being much more detailed and a lot more process and procedure than what I currently do - which is more akin to the version 0 described, but in parallel - we come up at the same final problem: reviews and quality assurance.
I sign off the code I merged, part of company policy but also just to be sure it is actually decent. But reviewing has become the real draining bottleneck: even stacked PRs, if that total 5-6k lines is not a 5min job. Even if I brainstormed and set the plan, that's really the part that doesn't scale right now for me in this. But the author is very shy about that: either the changes arent that big in the end or they trust the process enough to review in a more casual manner. Being equally untrusting I can't do that ...
> I don't think theres too much here I haven't seen elsewhere.
Isn't that the rhyme here. I can't think of any article or discussion on AI here that contains anything new or noteworthy. And yet all those articles we've read before and all those "discussions" we've had before keep coming and coming. I have gotten bored and I'm just waiting for anything decisive to happen.
I am completely calm regarding AI and development.
First nobody sane want to give their domain IP to OpenAI/Anthropic. That's why local AI will eventually prevail and flourish because people who actually have some IP will have no problem to buy 10k+ EUR machine to run some pretty good models on it. However if your main job is just doing CRUD stuff, then you are screwed.
Secondly hallucination is really Achilles heel of every LLM. Sure you can recreate an application which exists in thousand of variations on the internet, but the moment you will try to go more into domain knowledge you will start struggling more and more.
Try to make CAN driver for ESP32, easy it is probably going to work. Try to make CAN driver for STM32F7xx now the AI will start having a problem but probably will be able to produce something what is working after a lot of debugging. Now let's make CAN driver for MPC5555. AI will start writing fairy tales about registers which do not exist. All of processor above have reference manuals and sometimes example git repositories available on open internet.
Did you try this by giving it access to the materials? Human programmers also don't memorize all this stuff. If this is the reason for your calmness it's quite shortsighted.
There are problems when you rely too much on AI generated code, but these shallow dismissals are quite annoying.
> First nobody sane want to give their domain IP to OpenAI/Anthropic. That's why local AI will eventually prevail and flourish because people who actually have some IP will have no problem to buy 10k+ EUR machine to run some pretty good models on it. However if your main job is just doing CRUD stuff, then you are screwed
Replace OpenAI/Anthropic with AWS and this is not too dissimilar to the arguments in 2009 about cloud providers.
It’s not that there's nobody for whom this is true, it’s just that there’s enough of everyone else to build an empire with.
Good writeup. I think the main difference in my workflow is that I skipped the sandboxing part and accepted the coding agent having access to the entire 24/7 dev machine, so I'm still running on worktrees. Also, the "idea enrich" steps in my workflow are less formal - I tend to write most details in a feature spec myself. I also do my workflow on my own self-hosted custom interface which comes with a kanban board for project tracking, so I don't need Github. The rest of the workflow looks pretty similar.
*siiiighhh... Slop automation. Removing self from loop, automating brainstorming. It's madness. No way that code is any good, shippable beyond 2 users or even maintainable beyond auto-slapping on more slop. Sad.
"Automating Myself Out of Development" .. "I’m neither an AI-fanatic"
Clicks through bio, clicks into their linkedin .. wow, who could have ever guessed .. they're the CTO at an AI company whose sole product is trying to replace workers with AI solutions. Who could have ever guessed? That the person writing a personal blog post about how they're not an AI fanatic and how they can automate their work using AI tools .. is the CTO at an AI company selling products that do exactly that? Wow, what a coincidence. Surely this blog post contains a purely objective, impartial analysis of the topic.
I'm not some evil spy trying to make everyone use AI... both the startup and the blogpost are an experiment, a work in progress that I'm sharing I don't think anyone would read my blog and be like "Oh, now I need a tool for my SRE to automate incident management"
People are starved for real life examples, this is a real life example.
of course I'm not objective. To be objective you'd need to make a large scale study among thousands of companies. I'm sorry I don't have resources for that... just trying to get other people's feedback
30 comments
[ 2.7 ms ] story [ 72.8 ms ] threadi would not want to go down the "take myself out of the loop" path because yes, i do have to micromanage the claude session, often course-correcting every commit and then doing large scale refactoring every so often. but i'm perfectly happy doing that - i see claude as more of a tool than a coder i can hand work off to.
1. We've done it by hand for another route already, which the LLM uses as reference
2. Theres a strong validation setup/harness I've setup for it with storybooks, and component tests
3. It's a _mostly_ mechanical transform. Not entirely, as the two environments/APIs are not 1:1, but it's close enough
But! I and my team are still reviewing everything shrug it is "faster" because I get to have this running while I'm in meetings planning other more interesting projects
And this isn't really that many agents in parallel. Yeah, plenty of fan-out subagents, but that IMO doesn't count/isn't really the same as what others are talking about
The pipelines and data serving design was all human since it did have to deal with some data scale but the javascript/api layer was all slop, and it seems fine and good.
If you have a really high quality piece of code that needs to meet a high bar of quality/reliability, then I think the risk of letting the AI loose on it is very high and I wouldn't do it. If you have a pile of code you already know is a pile of garbage despite being human written, well, it can't get much worse :)
I also built an agent orchestration meta harness that runs on k8s and uses the k8s agents sandbox for running codex/claude code in the cloud. This was almost entirely just handed over to Fable and I have not asked a single architectural detail. The quality of this product is mediocre, but the fact that it largely works after I went through a few iterations of clicking around is impressive. I would have preferred to buy something off the shelf, but nothing even really came close (though maybe now I would have forked Omnigent)
My current process is also using Github projects in a normal scrum style way, with many tickets written or fleshed out and state managed by the LLM, and it doubling as the memory system
Completely leapfrogging all these other open and closed source concoctions and being more effective
But its effective enough that I don’t need OP’s final form state of still approving everything
Auto-mode is fine. Worktrees are built into Claude Code now. I just tell it to classify tickets as sequential or parallel possible and spawn subagents to tackle all of the tickets in the todo list
They all get their own context window its pretty perfect now
in the meantime I work in a couple tabs of Claude Design for different flows of any client side app. My philosophy has been that devs could pick up graphic and UI/UX design easily, its just still a full time job to make variations of layouts and portray their states.
UI/UX is not a full time job anymore.
And I use Claude chat to flesh out aspects of the overall idea
I think you may be overcomplicating your workflow in the concluding state.
Overall I agree that planning and intention is now most of the time, before a 10 subagent precision strike is initiated
It is not control freak behavior to want to be in control when you are the one accountable for it if it breaks.
I shudder when I hear about some people's (wildly overcomplicated) setups. I get the allure but there's something nice about pair programming with an LLM in a singular chat.
I sign off the code I merged, part of company policy but also just to be sure it is actually decent. But reviewing has become the real draining bottleneck: even stacked PRs, if that total 5-6k lines is not a 5min job. Even if I brainstormed and set the plan, that's really the part that doesn't scale right now for me in this. But the author is very shy about that: either the changes arent that big in the end or they trust the process enough to review in a more casual manner. Being equally untrusting I can't do that ...
>I want to start by saying that I’m neither an AI-fanatic
Kind of like saying you are a fanatic before saying you aren't.
I don't think theres too much here (e.g. "spec driven development") I haven't seen elsewhere.
Isn't that the rhyme here. I can't think of any article or discussion on AI here that contains anything new or noteworthy. And yet all those articles we've read before and all those "discussions" we've had before keep coming and coming. I have gotten bored and I'm just waiting for anything decisive to happen.
First nobody sane want to give their domain IP to OpenAI/Anthropic. That's why local AI will eventually prevail and flourish because people who actually have some IP will have no problem to buy 10k+ EUR machine to run some pretty good models on it. However if your main job is just doing CRUD stuff, then you are screwed.
Secondly hallucination is really Achilles heel of every LLM. Sure you can recreate an application which exists in thousand of variations on the internet, but the moment you will try to go more into domain knowledge you will start struggling more and more.
Try to make CAN driver for ESP32, easy it is probably going to work. Try to make CAN driver for STM32F7xx now the AI will start having a problem but probably will be able to produce something what is working after a lot of debugging. Now let's make CAN driver for MPC5555. AI will start writing fairy tales about registers which do not exist. All of processor above have reference manuals and sometimes example git repositories available on open internet.
There are problems when you rely too much on AI generated code, but these shallow dismissals are quite annoying.
Replace OpenAI/Anthropic with AWS and this is not too dissimilar to the arguments in 2009 about cloud providers.
It’s not that there's nobody for whom this is true, it’s just that there’s enough of everyone else to build an empire with.
Pretty much the whole industry has zero problem giving OpenAI/Anthropic full access to their systems and codebases.
You're putting way more thoughts into it than the vast majority, most companies seem to go with the momentum
Clicks through bio, clicks into their linkedin .. wow, who could have ever guessed .. they're the CTO at an AI company whose sole product is trying to replace workers with AI solutions. Who could have ever guessed? That the person writing a personal blog post about how they're not an AI fanatic and how they can automate their work using AI tools .. is the CTO at an AI company selling products that do exactly that? Wow, what a coincidence. Surely this blog post contains a purely objective, impartial analysis of the topic.
Nobody is entirely correct/wrong about the coming Future changes. Nobody knows what is reasonably believable, anymore...