Another Claude related article. It's starting to feel like spam now.
The creator of Claude Code's Claude setup
351–360 of 423 posts
Re: The creator of Claude Code's Claude setup
#352Earlier quoted context omitted.
> Can't speak for anyone else, but I personally know 3. I'm not saying your three friends/acquaintances don't exist, I'm saying the evidence suggests they aren't representative of a trend. This is consistent with the other evidence we have (e.g. studies which show that LLMs produce at best relatively modest gains in productivity, not enough for a one person team to do the work of even two people. > I'd also point out…
I'm not saying it's evidence for some larger trend; I'm presenting the reason why single-person teams might not advertise why they are single person and that these teams are not necessarily starting as single person teams, but sometimes collapsing down to single person teams.
Maybe you don't mean to, but when you present an anecdote you're implying evidence of some trend, otherwise it's just a pointless statement. And unless a multi-person team is collapsing down into multiple single person teams, there's no increase in productivity and we're actually in a worse position as a whole.
Re: The creator of Claude Code's Claude setup
#353This is interesting to hear, but I don't understand how this workflow actually works. I don't need 10 parallel agents making 50-100 PRs a week, I need 1 agent that successfully solves the most important problem. I don't understand how you can generate requirements quicky enough to have 10 parallel agents chewing away at meaningful work. I don't understand how you can have any meaningful supervising role over 10 thing…
I suppose he may have a list of feature requests and bug reports to work on, but it does seem a bit odd from a human perspective to want to work on 5 or more things literally in parallel, unless they are all so simple that there is no cognitive load and context switching required to mentally juggle them.
Washing dishes in parallel with laundry and cleaning is of course easily possible, but precisely because there is no cognitive load involved. When the washing machine stops you can interrupt what you are doing to load clothes into the drier, then go back to cleaning/whatever. Software development for anything non-trivial obviously has a much higher task-switching overhead. Optimal flow for a purely human developer is to "load context" at the beginning of the day, then remain in flow-state without interruptions.
The cynical part of me can't also help but wonder if Cherny/Anthopic aren't just advocating token-maxxing!
Re: The creator of Claude Code's Claude setup
#354Earlier quoted context omitted.
But eventually people will catch up you can basically create a working product alone with the help of AI. My prediction is that this will lead to a margin free-fall for many software products where the main moat is the software itself. And a lot of SaaS companies will also become redundant when the AI can code up a tailored solution in an hour for free.
This is mostly correct IMO. SaaS is extremely vulnerable, companies will be able to modify open source tools to do exactly what they need, and agents will make managing those services easier. This will lead to downward pressure on SaaS prices, and cause them to become more like cloud data management platforms that they let customers build on top of rather than one-size-fits-all apps.
I agree with this completely. I forsee an era of enterprise level 'template' saas products that are expected to be tinkered with and highly customized. I think products like Notion that have an incredibly robust customizability and integration layer are going to thrive, where every single company can use a template engine to build extremely customized applications - and the barrier to building on top of these will essentially become the rate of human speech.
Re: The creator of Claude Code's Claude setup
#355Earlier quoted context omitted.
Can't speak for anyone else, but I personally know 3. 2 of the 3 existed as entities for more than a year already, but pivoted at least once (both were VC-funded but now doing something very different than what they started with when I first met founders) and ultimately let go of their offshore and contract engineers once AI became good enough some time early last year. Founders basically realized that the quality of…
This matches what I’ve been seeing as well. Small teams can move surprisingly fast now, but the bottleneck usually shifts from engineering to distribution and positioning. We’ve found that building the product got easier, but turning it into a sustainable business still required just as much manual effort around sales, onboarding, and retention.
You're moving the goalposts; building the product never equaled writing some code, it's always involved all of the efforts you reference. The expectation is that you optimized the code generation and shifted the bottleneck, but are overall more productive (i.e. the cycle is shorter). If you're not iterating faster then there's be no productivity gain.
Re: The creator of Claude Code's Claude setup
#356This is interesting to hear, but I don't understand how this workflow actually works. I don't need 10 parallel agents making 50-100 PRs a week, I need 1 agent that successfully solves the most important problem. I don't understand how you can generate requirements quicky enough to have 10 parallel agents chewing away at meaningful work. I don't understand how you can have any meaningful supervising role over 10 thing…
Potentially, a lot of that isn't just code generation, it *is* requirements gathering, design iteration, analysis, debugging, etc.
I've been using CC for non-programming tasks and its been pretty successful so far, at least for personal projects (bordering on the edge of non-trivial). For instance, I'll get a 'designer' agent coming up with spec, and a 'design-critic' to challenge the design and make the original agent defend their choices. They can ask open questions after each round and I'll provide human feedback. After a few rounds of this, we whittle it down to a decent spec and try it out after handing it off to a coding agent.
Another example from work: I fired off some code analysis to an agent with the goal of creating integration tests, and then ran a set of spec reviewers in parallel to check its work before creating the actual tickets.
My point is there are a lot of steps involved in the whole product development process and isn't just "ship production code". And we can reduce the ambiguity/hallucinations/sycophancy by creating validation/checkpoints (either tests, 'critic' agents to challenge designs/spec, or human QA/validation when appropriate)
The end game of this approach is you have dozens or hundreds of agents running via some kind of orchestrator churning through a backlog that is combination human + AI generated, and the system posts questions to the human user(s) to gather feedback. The human spends most of the time doing high-level design/validation and answering open questions.
You definitely incur some cognitive debt and risk it doing something you don't want, but thats part of the fun for me (assuming it doesn't kill my AI bill).
Re: The creator of Claude Code's Claude setup
#357Earlier quoted context omitted.
But eventually people will catch up you can basically create a working product alone with the help of AI. My prediction is that this will lead to a margin free-fall for many software products where the main moat is the software itself. And a lot of SaaS companies will also become redundant when the AI can code up a tailored solution in an hour for free.
I think so too. But in the meantime there is a quiet goldrush for people who spot niches where they can extract decent (or a lot) of value right now, and for long enough to be worthwhile. If they can get scale enough that thinner margins makes for a worthwhile business when the market catches up, great. If they can't, then we stay lean we might make off with decent ROI. But that is also a reason to be cautious of cha…
> A lot of tech companies will find the institutional knowledge they thought would shore up their moat is worth a lot less than they thought.
I totally agree. I think going forward the primary value of SAS will be the embedded domain expertise in a pre-built product. The comparison of Asana versus Notion comes to mind for project management. Asana forces abstractions of good project management upon you, whereas Notion lets you build it yourself. I think this principle will scale to all software in the future, where the only real value of software or it becomes exported maintenance obligations and a predetermines domain abstraction.
But as you mentioned, I think companies will rapidly find that their own specific abstraction is worth a lot less than they believed.
Re: The creator of Claude Code's Claude setup
#358I feel like it's time for me to hang up this career. Prompting is boring, and doing it 5 times at once is just annoying multitasking. I know I'm mostly in it for the money, but at least there used to be a feeling of accomplishment sometimes. Now it's like, whose accomplishment is it?
you found meaning in the work vs the outcome. You can find meaning in the outcome with a new form of work.
Re: The creator of Claude Code's Claude setup
#359Earlier quoted context omitted.
You aren't just buying software, you're offloading liability of continued support and functionality.
Sure, but there's a whole lot of businesses already using custom solutions made with excel/access/etc that are held together with duct tape and chicken wire, so I think the adventurous spirit necessary is there.
In the same pattern there are a lot of businesses where these solutions are not efficient and they MOVED from them to expensive commercial software. It's actually an antipattern to build a bunch of in-house, Excel-based solutions - with AI or not - for these companies.
Re: The creator of Claude Code's Claude setup
#360Earlier quoted context omitted.
Most startup -> enterprise deals are like this in nature. Enterprise buyers are already wary of small startups (for various reasons). A 1 person startup? Wouldn't even get a meeting with the buyers in many cases even if your software was 10x cheaper and exactly solved the business problem.
I worked for a public health care Enterprise early in my career and I make a joke to one of the VPs once about how it seemed like the real career success would be finding one of our pain points as a patient or employee, leaving to start a company that solves that, and selling it back to us. He laughed and said several people had done that but you better take a half dozen executives with you or you'd never get the fir…
> you better take a half dozen executives with you or you'd never get the first meeting no matter how good the product was
In my specific space, clinical trials can run for years. A company that might fold if they run out of runway? Non-starter. I was a member of a small company that did make this work and it required that we put our code in escrow with a large multi-national IT company that owned the support contract (customer paid us for licensing, paid multi-national IT company for support, our source code went into escrow).