This isn't about how to build an AI-native startup, it's how to use Anthropic tools to automate 2019 style app building. An AI native startup would have AI infused into the product, and Anthropic doesn't want anyone but them to sell AI.
I'm interested in the concrete technical stack that "AI-native" start-ups use. No engineers and founders just go to full production scale on something like Lovable? Or code in GitHub, with something like Jules driving all the development via GitHub issues and comments (sidenote: does anyone have a good workflow for doing this with Claude/GPT?).
As someone who was the technical co-founder of a company before AI, I really feel like a lot of that role can be done by AI, if the foundation is laid appropriately.
It's nice to see an explanation of how they think you should use Claude for a host of different job functions / aspects of building a business, but the tone makes it seem like founding a startup is something you wake up one morning and just decide to do instead of, say, going to the park. Over coffee, you ask Claude about your idea and when it tells you "you're a genius" you're off. That's silly in so many ways. Statements like this "Validation cycles that used to take months now take afternoons" have an element of truth but ring with false promise.
And that relates to the lack of timelines and focus on how long things took around 2020 BC, that is Before Claude. Building a startup isn't like having a lemonade stand as a kid where you just don't bother to do it if you forget to buy lemons or it's rainy or something more fun comes along. There's a significant compound interest element to startups that's easy to overlook. Your codebase grows over time and so does your feature set and that collection of features attracts customers in a way one thing might not. You learn as go, of course, too.
This seems particularly relevant to the GTM section, which I was particularly amused by since that's what I'm focused on right now. It's a long game. Your blog post doesn't get found by anyone in Google until you've built up your SEO mojo, your LinkedIn post isn't read without the followers you need to accumulate and your content has to get engagement for people to see it even then, you don't start off line with a million followers on X, etc.
> Your blog post doesn't get found by anyone in Google until you've built up your SEO mojo, your LinkedIn post isn't read without the followers you need to accumulate and your content has to get engagement for people to see it even then, you don't start off line with a million followers on X, etc.
I hate that this is true. It's the worst part about selling stuff online IMO and I found that you have to spend so much time doing it. In many cases, selling something online can be optimized to the extreme such that spend on marketing should be as high as possible and spend on the product R&D, manufacturing, support, etc should be minimized as much as possible. This equation gives you the most profit, but also gives the customer the absolute worst product that is possible to sell.
Capitalism doesn't really have a solution to this problem that I've seen yet.
> the tone makes it seem like founding a startup is something you wake up one morning and just decide to do instead of, say, going to the park.
I'm not sure it's a crazy idea when you can run a whole revenue generating company with basically zero employees. You could have a successful 'startup' generating 200k in revenue a year, you just need to cover the cost of your anthropic subscription.
>As an AI-native startup founder, your responsibility is to know what's in your
codebase, understand any potential exposure vectors, and not ship obvious
vulnerabilities to real users who are trusting you with their data.
This is fairly funny coming from the company whose employees report merging in hundreds of PRs per engineer per day, and accidentally leaked their own source code through a security misconfiguration in a package manager they own.
> your responsibility is to know what's in your codebase, understand any potential exposure vectors, and not ship obvious vulnerabilities to real users
It seems like CYA; with all the marketing about how LLMs will solve all problems it was really surprising to see that, but legal probably told them to go easy on it.
Hundreds of PR’s per engineer per day! They would have zero visibility of their code. Their AI’s would have no visibility of the million plus lines of code.
100 feels low given I just saw dependabot do 8 in one hour. No AI required!
It matters a lot what size the PRs are, and this varies wildly from place to place. I spoke to someone who instituted a “no PRs over 500 lines” rule. I would refuse to even read something that big unless it was just a find and replace or boilerplate.
It's a slide deck telling people what a product can do (that's a normal thing to release for a company), but the thing it tries to sell you on is building your own business based on their tool.
Which makes no sense the way they sell it, because "founding a business" is no standard process that could be formalized in a way like that, nor does it make sense for society to have people founding businesses at a scale comparable with mowing your lawn or doing your taxes.
All of this feels just unreal because it is unreal. Founding cannot be a commodity. If it is, you have no moat or point, meaning you instantly collapse again, because you are an interchangeable commodity.
It makes sense if you think of 'founder' as an identity like 'influencer', rather than 'someone actually seriously founding a business'; just as with influencers, some people will make a lot of money doing it for real, but many, many more will post enthusiastically on social media, living the aesthetic.
A lot of people already treat being a founder/entrepreneur as who they are, not what they do--witness the endless tide of LinkedIn posting about hustle culture, divorced from reality. This is an extension of startup chic.
"Founding cannot be a commodity. If it is, you have no moat or point, meaning you instantly collapse again, because you are an interchangeable commodity."
IMHO you still need to find the product and PMF
There are bunch of books startup world recommends which sort of all start from the principle of product, users, traction.
This is sort of scaffolding around that. It's not entirely insane to try to formalize this process - there already are books that do this (Bill Aulet, Disciplined entrepreneurship).
"nor does it make sense for society to have people founding businesses at a scale"
Maybe not at scale of moving lawns but I'm pretty sure the world is full of nichces that still lack specific software offering or where options of software offerings are limited.
This is like "Uber for logging" or "time reservation system for cat dentists" level of "take existing product category and apply to a domain you know".
So not every cat dentist needs to found a cat dentist time reservation app but I'm sure there are niches withing niches with business opportunities awaiting.
Yeah, and these days it really isn't a big deal to build things; it's much bigger challenge to actually develop a distribution channels and cut through the noise. I think people are just overwhelmed with everything and attention span is shorter and shorter. And that's the real issue - what I am finding now is that again the thing that really works is good old actual human conversation with potential clients.
> Founding cannot be a commodity. If it is, you have no moat or point, meaning you instantly collapse again, because you are an interchangeable commodity.
With the recent AMD announcement [1], local models are likely the future indeed. Cloud will be the place for remote sessions, remote agents, continuous agents. But I foresee a place where phones, laptops, PCs, and even perhaps dedicated hardware just for AI, will be the place for most AI-related workloads.
Their argument on page 10 is that now agentic coding reduces the effort of writing code there will be far more failures unless you validate the idea properly.
We are actually seeing that in that the number of apps on the app stores is increasing but usage is not increasing.
Some would argue that the right process will lead to the right results.
> there will be far more failures unless you validate the idea properly
You are exactly correct! You're very insightful to see that AI is perfect for validating ideas. It's not just sucking up, it's swallowing. Would you like to delve into coding up a Synchopancy as a Service web site?
The barrier to entry drops and so more garbage enters the market. This has happened many many times but there can be a massive and beautiful paradigm shift at the same time. Think about YouTube. Most stuff on there is garbage, but compare the good stuff of today to pre-YouTube content. To be quite honest, I think most of the best media is going straight onto YouTube.
I think a lot of founding is pretty much a commodity, e.g coming up with a viable idea and then implementing it has become rather easy now with these tools. The real barriers are access to capital and clients. From the startup I joined (I'm the 6th person) I see how much the founders personal connections are important. That indeed can't be commoditized yet. But the process of coming up with an idea and iterating on it ? The founders didn't even come up with our idea - they thought of something initially but the investor led them to his own idea - totally different. That's how the company was born. Now the first clients are connected to the investors. Etc.
So access to capital and clients, connections ,that's the last standing moat I think.
It's a commodity in the same way that making music is a commodity (i.e. using production tools to make it sound good). But music today is so much more generic and boring than it used to be.
It’s reasonable to say that there are things like discovery and validation that are necessary for every startup (to succeed) and that some techniques for these can be automated.
That we can describe something like “validation” in the abstract and automate some part of it says nothing about whether it’s worthwhile. I’m hard pressed to think of anything that anyone pays for that doesn’t meet this description. Why should being a founder be special or different?
>nor does it make sense for society to have people founding businesses at a scale comparable with mowing your lawn or doing your taxes
It absolutely does. AI and robots drives the cost of labour down; it's good for capital, bad for labor. If everyone is a business owner then everyone can benefit. A hundred years ago the majority of Americans were self-employed; mass wage labor is a recent phenomenon.
If that’s what’s actually going to happen. If people start businesses entirely dependent on Anthropic’s ecosystem, they’re more akin to drivers for Uber who basically are employees now stripped of employee wage protection.
If it’s a guy starting his own taxi service with his own app he paid a guy $2000 in Ukraine to produce (as a guy where I live did), that’s different.
A beautiful analogy for non-technical founders creating software products with AI. There are version control systems, but who needs them when you can name your pdf `n-final-updated-6-16-final-donottouch.pdf (1)(2)`
I’m pretty sure the one place people will never believe AI can be applied is “being a founder”.
There’s just too much invested, in terms of beliefs and money into the idea that founders are special and therefore deserve seven-eight figures off of the capital pumped into their unprofitable products.
You’ll see it here in comments. People will defend A”I” applied to software engineering wherever (not) possible, but building companies? Now listen buddy there’s an irreplaceable human genius at work.
Exactly, I always find it ridiculous how the suits, any layer of mid-managment to executives, are so eager on AI 'outsourcing' everything, but they themselves think the 'outsourcing' (if it really works) would stop just before their position.
Anecdotally, Someone I know said that their manager just asks them if they have drank water or not and motivate them, and the company wants them to make on their personal time what they are building with AI and showcase it.
So they were joking that they were building a replacement of their manager using Claude, and although they were joking, but only partially.
If anything, the managers, rationally speaking might be the one to be outsourced.
Of course, though recently I have come to realize that world isn't completely rational but I have found that on the long term, it still rewards for rational behaviour (sometimes) so I think that these managers on Linkedin should feel a bit scared... (as they are probably just prompting AI to write the linkedin-texts)
What if they are already scared and this is their way of coping with it? Nobody wants their job to be redundant and I wouldn't say that tech is completely scott free but rationally I find that tech has this taste/authenticity factor and there are too many factors but I must admit that these AI companies are succeeding in trying to force the narrative about jobs being completely redundant combined with job market, to make the labour vs capital divide even larger.
I feel as if this is why people want to be on that side of the battle and this Linkedin-Ai-founder-syndrome is a symptom not the cause.
We engineers are trying to optimize from the tech side through let's say AI to building lots of tools that we otherwise couldn't have had made due to lack of time but we had many ideas from any issues we face at tinkering/work etc..
and the managers not having to do any of it, lack the ideas in that aspect but they succeed in trying to project something even if the reality of that thing doesn't exist. So they are trying to do that.
I don't know, I don't believe much in the divide of engineer vs manager but the fight is larger than it but we are infighting...
We are all just fighting to get to the other side of the line of a broken system but oh well, so it is and I am unsure in what capacity can we fix it but I still hope deep down that things will improve for better because I guess hope is what makes us human.
Why would that be true? Successful founders have to be unsentimental by nature. If you make it harder on yourself than it has to be, you just get killed by people who don't.
I demo'd an automated agent platform here and showed it doing PRs, and the CEO asked me what i'd do for a living after this did all my work for me, and I said, "What do you think that you do that an agent can't do?".
While you're not wrong the problem is they own the company or the board listens to them. They have power they can use to keep existing while you don't. This is why AI is so scary to working professionals.
AI has changed the build for sure, it is a lot easier to build now, a lot easier to practice multiple copywriting ideas, do market research ... etc.
There is something that will never change for being a founder, you need to sell, and for that you need network and credibility. It was never about the building, its all about the selling. AI has not changed that.
But it has. AI can help you do market research, develop buyer personas, evaluate potential customers, create, analyze and enrich prospect lists, evaluate marketing channels, create ad copy, write sales scripts, think through objections and how to respond, etc.
Will it turn you into Jordan Belfort? No. Will it be 100% successful or effective? No. But can it help enough to make a difference? Sure, in enough cases.
Assumption: now everyone can do more of the above. The final line is still selling. So everyone will get to the sales part, FASTER. Triage will still happen at this stage, regardless of AI. You won’t be able to avoid this triage, regardless of how fast you get there.
I can't find a name to dig more but the "everyone will get" part is something strange to me. If everybody has the same capability increase, then what changed really ? some would even say it will increase the paradox of choice.. more offer, still the same amount of time to decide, or maybe more AI based decision to match the amount.. so less human understanding.
“everyone”. it’s there, it’s accessible, it’s “cheap”. acceleration will depend on the operator capability. if the final product will make a diff in the real world, it will ALWAYS depend on the entrepreneur, not the tools used.
The difference is the destination. Me crunching code with llms vs you doing the same thing will lead both of us to different places(let’s say equally fast for the sake of the conversation). What will set us apart in the end is who will actually sell it, and that’s llm-independent.
Unfortunately it feels close to zero sum to me. I am getting absolutely drowned in AI generated personal sales pitches now. That obviously scraped my name/company online and automated the email. I feel sales becomes even more relient on trade shows and conferences and person to person interaction (Only talking about B2B stuff that I am involved in).
If anything, AI has made it more difficult and challenging because every customer and investor is drowning in AI-generated collaterals, websites, etc. The situation is dire in the academic world, where both the applicants and the reviewers now rely so heavily on AI that both publishing and financing has turned into a lottery.
I am positive this will settle down at some point, but the difference will always remain about your own abilities, not that of AI.
> If anything, AI has made it more difficult and challenging because every customer and investor is drowning in AI-generated collaterals, websites, etc.
In many markets, yes. If you're a software buyer, for example, your inbox, LinkedIn, etc. is filled with AI-generated sales outreach. And you know it's AI.
But keep in mind that there are tons of markets (think local services) where buyers aren't familiar with AI. They don't know that what they're reading was produced by AI, and they wouldn't care.
In these markets, if you use AI, you have a realistic shot at being "better" than your competition, and if you use it even a little bit more effectively, it can make a real difference.
I get it, but it is still more difficult to achieve differentiation from your less-skilled competitor in the short term, because they can simply slop their way through, at least until prospects realize that this is a bag of sh%t
The same thing happened in other businesses where it became easier than ever to put out content or product, but harder to break through because of all the saturation.
If you're using AI for your marketing you're going to get lumped into a slop category, with plenty of other products to keep you company. Only people with AI psychosis actually believe this garbage. All LLM output has a cheap stench to it that's impossible to ignore.
There is no shortcut to hardwork, but llms somehow have people thinking that is the cases, it plays so well into people's desire to be as lazy as possible.
Outside of the tech/AI bubble, you'd be amazed at how few people can spot AI-generated content, and how few people seem to care if they think the AI-generated content speaks to their needs.
I know a small business that generates many of their leads by responding to posts on social media. They recently started using AI to create personalized comments responding to these posts instead of generic comment templates they used before. The number of leads they're generating from their social media commenting has skyrocketed.
comments yes, because people have their guard down. Its just lazy. Good for them I guess, until they lose their accounts or someone like me notices and puts them on blast.
Nobody cares. That's the thing. If you're using AI in a relatively "thoughtful" way to more efficiently provide information that's of value, the average person isn't concerning themselves with the fact the end product was produced by AI.
Facebook, Instagram, LinkedIn are flooded with AI content. A lot of it is "slop", yes, and it's kind of annoying if you know where it came from, but you're missing the big picture if you deny the reality, which is that if AI content is created in a way where it does have some value, average people appreciate the value and don't have some ideological hang-up over how it was created.
AI is just a tool. It can produce absolute rubbish, but it can also produce some decent stuff too.
With all due respect this reads a little deranged. To sell something to the masses you fundamentally need a product to sell. I'll agree that how you market the product can be a "product" in itself, but that only gets you so far. If it was never not about the building why waterfall vs agile why velocity why stakeholder why business analysts why meetings why board members pushing for features?
This is like when AI bros claim that AI has changed absolutely everything for their project but the first thing they do is reach for docker compose, react and postgres. Why don't you forget the bloat and have your LLM make your container, vdom differ and lightweight DB?
Especially as someone outside the US, building a startup on AI sounds like a bad idea. Some AI company fails to pay their bribes on time, or your country doesn't cede territory to the US president, the AI gets yoinked and you are left with Mistral or Qwen.
(Technically that also applies to MS Teams, Google and so on and not just AI)
> Now someone with no engineering background can build production
software that brings their idea to life, while a technically adept founder with little business knowledge can easily produce a go-to-market strategy, a financial model, and a highly polished pitch deck.
I had a bit of a laugh. The non-technical business experts are much more likely to achieve success than the technical experts. They can actually talk to the customer and get the customer to care. No quantity of GPUs and gas turbines can correct for a lack of personality or reputation. The technology is generally not the hard part in most businesses, despite the extreme efforts of certain technology people to make it seem so.
Does this include making annoying Linkedin posts every other day about how AI 30x'ed your engineering output and killed graphic design for real this time?
> The traditional startup growth arc assumes that the path from idea to scale
is validate → raise → hire → build → raise again → grow → hire more → repeat.
Now, AI has erased the expectation that each new phase in the startup lifecycle
requires a bigger team, a different skill set, and a fresh funding round.
From 2015-2019 I spent the whole time saying "If I don't write the code nobody does". It was the point of saying to do anything requires a team, to build that team you need funding. It was a vicious cycle and took a long time to get enough traction to raise funding and do that... and then you end up in the MVP loop -> hire -> build -> validate -> rehire -> rebuild -> revalidate. Today all of that has changed. You don't necessarily need the team to write the code, it's for a different function entirely, maybe the original function which is the team was the orchestration engine for all the different pieces at play to make a company and product successful. The code is only half the equation. Looking forward to seeing how solo entrepreneurs leverage these tools and how teams transform using them.
Feels like the people who want to do it spend their time here arguing about it, and the people who will do it are the ones in the trenches epxerimenting and trying to make it work. The OP is just insight porn at this point.
This framework does not sound that much different from what people used to do but with AI agents and coding assistants, and this is not going to work unless you are lucky. LLMs cannot come up with good ideas and coding (if you believe it is solved) is no longer moat especially in the early stages.
So either go viral or go home.
Obviously personal connections, timing, market position all play role but let's be honest - this is not something that can be planned although in retrospect it may seem so. A % of the population will get all of this right many times in a roll but this is just mathematical certainty.
A serious AI-native founder should not waste time reading this brochures, they should make agentic loops instead where their agents autonomously read and produce brochures for their brochure first, product second startups.
There is still so much old school thinking and process in this. Go through this process that has been the same for the last decade+ but now you just don't need a team of people to do it. You can just use Claude instead! Really, we are in a paradigm shift, not in a "do the same thing but with less people" shift.
The challenge: Ask an AI tool for evidence supporting what you already believe,
and it will find it. Confirmation bias now comes with a research engine.
I just passed the idea stage and then hit the MVP stage with a prototype with a potential customer who's definitely said they would buy it!
And I'm noticing there's going to be a whole bunch of other things that I haven't thought of around actually running this thing and the timing is perfect for this playbook!
Good guide but I think the product market fit portion of a startup is so key that you need no other skills except that to make a good startup. AI won't help you with that portion, only in depth knowledge of a industry or natural product intuition will.
Who knows, maybe an AI ideated and AI created product will be the best app of 2026.
I feel like a lot of this advice is kind of dangerous. How do I draft a tight investor memo? I'll ask the slop machine!
It's kind of analogous to how I'm writing code right now. For simple stuff or low priority stuff I'll fire claude at it and won't look at the code if it works. But for the important stuff I'm very carefully integrated into the cycle making sure what's coming out at the end is just right. I'm carefully constructing prompt loops and validation cycles to make sure what comes out looks like what I want - because I have the knowledge and experience of what works for my specific use case. Drafting an investor memo seems like the second category of thing, you need it to be right. I don't think claude offers much of value there. What's more - if you start slopping your investors, you are going to piss them off. Unless Claude is going to say it has some special data source it's used to train on so it knows good from bad, I think this is a bad idea.
This article also kind of fits in the category of "Here's how to use AI for EVERYTHING!" and actually it would be far more valuable to say "This is the bits that AI is good at, and here's where you need to do it yourself" - which is obviously a position that Anthropic can't hold.
I've noticed that seemingly every single tech company has re-branded themselves as "AI" company. Add a RAG system and you're now AI. Add a AI-chatbot, and you're now AI.
IMHO most founder will fail because leaning to heavy into ai and creating a system where they never experience the friction necessary to build the domain knowledge which ultimately could be the deciding factor.
Just think about website design, I don’t think it’s far-fetched to say that a non ai design website will outperform an ai designed one. These percentages add up in multiple disciplines.
I would argue betting against ai is your best chance of succeeding frankly (not in all cases but certainly as displayed here)
I think it's easy for those already in the tech industry to pooh-pooh this, as the previous comments on this post have.
Right now, people with ideas prompt their LLM by saying "I know how to make x, how do I turn that into a business?" Anthropic knows that, and releasing a playbook like this is a way to make people who haven't asked that question think to ask it.
For a non-technical person with a small business they don't know how to operationalize, an agentic workflow is a game changer. You might go from only getting 30% of your work time to build and improve your actual product to 50% or 70%.
Can you imagine having a knitting business, and suddenly being able to gauge interest for different colors with a website selector you'd have no idea how to automate? Or needing to close your shop for an upcoming holiday, and having Google and Apple Maps and your website all updated to reflect your closed dates cleanly, without having to fight through every UI? An engineer goes "bah", a baker goes "I just got to sleep two more hours".
I truly think that people in the tech industry do not understand how hard technology is for people who aren't in it.
> Can you imagine having a knitting business, and suddenly being able to gauge interest for different colors with a website selector you'd have no idea how to automate?
This does not sound like an issue small craft businesses have, but something programmer think is a thing.
I think you're thinking of the value in "creating a selector", which is not my point.
Businesses need to gauge interest and experiment with options. Knitting colors was my example of that, since I've got some personal experience helping a knitting business.
If you're a nontechnical craft business owner, the work of figuring out what questions to ask to grow your business is absolutely nonintuitive, and it's one of the things LLMs are best at. The fact that it can also modify your website or listing or what have you is an additional benefit; I chose to use that specific benefit to make the concept potentially more legible to the HN audience.
this is clearly targeting either vibecoders or people with extremely limited experience in software. i feel like it would be obvious to anyone who doesn't fall into those two categories that the 'AI wrapper' product category is both flooded and has a short shelf life. so many of these products are going to get wiped out once anthropic and other frontier labs stop pricing tokens below market and charge what they actually need for profitability. this is a short term cash grab at best. but hey, they gotta move tokens.
>this is clearly targeting either vibecoders or people with extremely limited experience in software
From the PDF:
>Launch energy is generated from ephemeral forces, like your founder’s friends, prospective buyers at your investor’s other portfolio companies, or a Hacker News headline that drives a spike. Unfortunately, none of these reliably predicts what happens at week six or week twelve when that initial boost has faded
I never understood the line of thinking so many people are trying to push. “Yeah you can build your own CRM now but the marketing and selling is the hard part.” Who said I wanted to sell it? I’m building it for myself so I can ditch SalesForce. Do I need to sell it to others for it to benefit me?
The cope here is that no one will build there own Shopify or SalesForce or Airtable because “the selling is the hard part.” They don’t need to sell and market it for those SaaS to fall.
Huh, I had to search this website-files.com website and it seems to be of webflow.
Was the PDF created by webflow? a bit ironic.
Maybe Anthropic should read their own founder's playbook to create a replacement to this tool.
(Yes, I know that when involving cdn's, there's much more to the infrastructure as well but isn't that what anthropic is promising in the founder's playbook that you can be a founder which solves problems [from my understanding],
My point is that there's probably much more to it than what might've been suggested which isn't a neutral source in the first place)
When I see notes like this, I wonder whether every success story can really be summarized and patternized this way. If you're building an AI based startup, what exactly would be the point of differentiation? That seems to be the difficult part
I've been at a few VC / startup events recently and I was stunned to see the number execs frothing at the mouth about finding a 1-person-ai-driven-billion-dollar-startup. This "playbook" is probably not going to help.
This isn't about how to build an AI-native startup, it's how to use Anthropic tools to automate 2019 style app building. An AI native startup would have AI infused into the product, and Anthropic doesn't want anyone but them to sell AI.
I'm interested in the concrete technical stack that "AI-native" start-ups use. No engineers and founders just go to full production scale on something like Lovable? Or code in GitHub, with something like Jules driving all the development via GitHub issues and comments (sidenote: does anyone have a good workflow for doing this with Claude/GPT?).
As someone who was the technical co-founder of a company before AI, I really feel like a lot of that role can be done by AI, if the foundation is laid appropriately.
It's nice to see an explanation of how they think you should use Claude for a host of different job functions / aspects of building a business, but the tone makes it seem like founding a startup is something you wake up one morning and just decide to do instead of, say, going to the park. Over coffee, you ask Claude about your idea and when it tells you "you're a genius" you're off. That's silly in so many ways. Statements like this "Validation cycles that used to take months now take afternoons" have an element of truth but ring with false promise.
And that relates to the lack of timelines and focus on how long things took around 2020 BC, that is Before Claude. Building a startup isn't like having a lemonade stand as a kid where you just don't bother to do it if you forget to buy lemons or it's rainy or something more fun comes along. There's a significant compound interest element to startups that's easy to overlook. Your codebase grows over time and so does your feature set and that collection of features attracts customers in a way one thing might not. You learn as go, of course, too.
This seems particularly relevant to the GTM section, which I was particularly amused by since that's what I'm focused on right now. It's a long game. Your blog post doesn't get found by anyone in Google until you've built up your SEO mojo, your LinkedIn post isn't read without the followers you need to accumulate and your content has to get engagement for people to see it even then, you don't start off line with a million followers on X, etc.
> 2020 BC, that is Before Claude
To avoid confusion and improve clarity, I propose 'BCC': Before Claude, Codex, et al.
Before Covid
> Your blog post doesn't get found by anyone in Google until you've built up your SEO mojo, your LinkedIn post isn't read without the followers you need to accumulate and your content has to get engagement for people to see it even then, you don't start off line with a million followers on X, etc.
I hate that this is true. It's the worst part about selling stuff online IMO and I found that you have to spend so much time doing it. In many cases, selling something online can be optimized to the extreme such that spend on marketing should be as high as possible and spend on the product R&D, manufacturing, support, etc should be minimized as much as possible. This equation gives you the most profit, but also gives the customer the absolute worst product that is possible to sell.
Capitalism doesn't really have a solution to this problem that I've seen yet.
> the tone makes it seem like founding a startup is something you wake up one morning and just decide to do instead of, say, going to the park.
I'm not sure it's a crazy idea when you can run a whole revenue generating company with basically zero employees. You could have a successful 'startup' generating 200k in revenue a year, you just need to cover the cost of your anthropic subscription.
>As an AI-native startup founder, your responsibility is to know what's in your codebase, understand any potential exposure vectors, and not ship obvious vulnerabilities to real users who are trusting you with their data.
This is fairly funny coming from the company whose employees report merging in hundreds of PRs per engineer per day, and accidentally leaked their own source code through a security misconfiguration in a package manager they own.
> your responsibility is to know what's in your codebase, understand any potential exposure vectors, and not ship obvious vulnerabilities to real users
It seems like CYA; with all the marketing about how LLMs will solve all problems it was really surprising to see that, but legal probably told them to go easy on it.
The best way to learn is from other people's mistakes.
Hundreds of PR’s per engineer per day! They would have zero visibility of their code. Their AI’s would have no visibility of the million plus lines of code.
Sounds super stable and cool.
Yeah this is a Mythos pitch.
100 PRs a day? I am sure this is hyperbole but otherwise you have a quote for me?
100 feels low given I just saw dependabot do 8 in one hour. No AI required!
It matters a lot what size the PRs are, and this varies wildly from place to place. I spoke to someone who instituted a “no PRs over 500 lines” rule. I would refuse to even read something that big unless it was just a find and replace or boilerplate.
Here's your quote:
"employees report merging in hundreds of PRs per engineer per day"
Feels like a category error.
It's a slide deck telling people what a product can do (that's a normal thing to release for a company), but the thing it tries to sell you on is building your own business based on their tool.
Which makes no sense the way they sell it, because "founding a business" is no standard process that could be formalized in a way like that, nor does it make sense for society to have people founding businesses at a scale comparable with mowing your lawn or doing your taxes.
All of this feels just unreal because it is unreal. Founding cannot be a commodity. If it is, you have no moat or point, meaning you instantly collapse again, because you are an interchangeable commodity.
It makes sense if you think of 'founder' as an identity like 'influencer', rather than 'someone actually seriously founding a business'; just as with influencers, some people will make a lot of money doing it for real, but many, many more will post enthusiastically on social media, living the aesthetic.
A lot of people already treat being a founder/entrepreneur as who they are, not what they do--witness the endless tide of LinkedIn posting about hustle culture, divorced from reality. This is an extension of startup chic.
this is well said. and to me this is crazy and sad at the same time.
An enormous amount of LinkedIn appears to be more about gatekeeping what other people can do than doing anything yourself.
They sell decorations, refreshment, cakes, costumes, games, and invitations for Founder Cosplay Parties.
"Founding cannot be a commodity. If it is, you have no moat or point, meaning you instantly collapse again, because you are an interchangeable commodity."
IMHO you still need to find the product and PMF
There are bunch of books startup world recommends which sort of all start from the principle of product, users, traction.
This is sort of scaffolding around that. It's not entirely insane to try to formalize this process - there already are books that do this (Bill Aulet, Disciplined entrepreneurship).
"nor does it make sense for society to have people founding businesses at a scale"
Maybe not at scale of moving lawns but I'm pretty sure the world is full of nichces that still lack specific software offering or where options of software offerings are limited.
This is like "Uber for logging" or "time reservation system for cat dentists" level of "take existing product category and apply to a domain you know".
So not every cat dentist needs to found a cat dentist time reservation app but I'm sure there are niches withing niches with business opportunities awaiting.
> Uber for logging
Lumberjack gig economy?
Interested.
Yeah, and these days it really isn't a big deal to build things; it's much bigger challenge to actually develop a distribution channels and cut through the noise. I think people are just overwhelmed with everything and attention span is shorter and shorter. And that's the real issue - what I am finding now is that again the thing that really works is good old actual human conversation with potential clients.
We have seen that category in the past, in MLM.
Perhaps now it’s only two levels but still somehow pyramid shaped.
> Which makes no sense the way they sell it, because "founding a business" is no standard process that could be formalized in a way like that
It makes no sense, but most technical people wished it could be like that and that's who this article is aimed at
Yes but that means that it is lying in people's face. Really just straight-up lying. No fancy words to soften it. Just lies.
Lying is bad.
As said in Seinfeld: "Jerry, just remember. Is not a lie, if you believe it"
How people cope with their delusion is irrelevant tho.
They're delusional. They must stop
> Founding cannot be a commodity. If it is, you have no moat or point, meaning you instantly collapse again, because you are an interchangeable commodity.
With the recent AMD announcement [1], local models are likely the future indeed. Cloud will be the place for remote sessions, remote agents, continuous agents. But I foresee a place where phones, laptops, PCs, and even perhaps dedicated hardware just for AI, will be the place for most AI-related workloads.
Their argument on page 10 is that now agentic coding reduces the effort of writing code there will be far more failures unless you validate the idea properly.
We are actually seeing that in that the number of apps on the app stores is increasing but usage is not increasing.
Some would argue that the right process will lead to the right results.
> there will be far more failures unless you validate the idea properly
You are exactly correct! You're very insightful to see that AI is perfect for validating ideas. It's not just sucking up, it's swallowing. Would you like to delve into coding up a Synchopancy as a Service web site?
Hah. Poe's law.
The barrier to entry drops and so more garbage enters the market. This has happened many many times but there can be a massive and beautiful paradigm shift at the same time. Think about YouTube. Most stuff on there is garbage, but compare the good stuff of today to pre-YouTube content. To be quite honest, I think most of the best media is going straight onto YouTube.
I think a lot of founding is pretty much a commodity, e.g coming up with a viable idea and then implementing it has become rather easy now with these tools. The real barriers are access to capital and clients. From the startup I joined (I'm the 6th person) I see how much the founders personal connections are important. That indeed can't be commoditized yet. But the process of coming up with an idea and iterating on it ? The founders didn't even come up with our idea - they thought of something initially but the investor led them to his own idea - totally different. That's how the company was born. Now the first clients are connected to the investors. Etc.
So access to capital and clients, connections ,that's the last standing moat I think.
It's a commodity in the same way that making music is a commodity (i.e. using production tools to make it sound good). But music today is so much more generic and boring than it used to be.
You seem to be mistaking the "rules" of the ZIRP SV nonsense bubble for the rules of reality.
Understandably so, but still.
It’s reasonable to say that there are things like discovery and validation that are necessary for every startup (to succeed) and that some techniques for these can be automated.
That we can describe something like “validation” in the abstract and automate some part of it says nothing about whether it’s worthwhile. I’m hard pressed to think of anything that anyone pays for that doesn’t meet this description. Why should being a founder be special or different?
Welcome to the world of companies founded by business school grads. No soul, no moat, just an endless cycle of KPIs and billion-dollar exits to PE.
Come on, get out of here with this “only true programmers can found a real company”
Such gatekeeping.
>nor does it make sense for society to have people founding businesses at a scale comparable with mowing your lawn or doing your taxes
It absolutely does. AI and robots drives the cost of labour down; it's good for capital, bad for labor. If everyone is a business owner then everyone can benefit. A hundred years ago the majority of Americans were self-employed; mass wage labor is a recent phenomenon.
If that’s what’s actually going to happen. If people start businesses entirely dependent on Anthropic’s ecosystem, they’re more akin to drivers for Uber who basically are employees now stripped of employee wage protection.
If it’s a guy starting his own taxi service with his own app he paid a guy $2000 in Ukraine to produce (as a guy where I live did), that’s different.
The same is true of building a startup based on Rails or React.
Maybe next we can get a blog post from Herman Miller telling you the “playbook” for how to run your startup on tables and chairs.
> "founding a business" is no standard process that could be formalized
Does experience founding a business make you better at founding another, unrelated business? I would say it does to some extent.
It often doesn’t. A common mistake of a successful founder is thinking they can repeat it. Those who can are exceedingly rare.
This feels like a “sell the shovels” move. Social media is full of “this one prompt to get rich quick”. It’s the new “one weird trick”.
Build a $1B startup. Make no mistakes. /s
"I will send you to jail if you fail" really makes it work all of the time
I know in which DC you are hosted and will cut your electricity off.
I like how dates and copies are still the ultimate version control:
"<filename>-05062026_v3 (1).pdf"
So there were 4 iterations on 5th of June alone for this document
A beautiful analogy for non-technical founders creating software products with AI. There are version control systems, but who needs them when you can name your pdf `n-final-updated-6-16-final-donottouch.pdf (1)(2)`
And a good file store and organiser is arranging all kinds of icons in different corners on your desktop.
This should be obvious but why would you trust what the spade seller says about being an AI-native startup.
Even if you believe AI-native startup is the future (the comments are divided), you would at least want to hear from an impartial source.
This is just marketing material.
AI has dramatically shortened the time from idea to product.
But for shortening the time from product to customer, you still need to figure that out
I looked at the PDF and confirm there is nothing of value in there.
I’m pretty sure the one place people will never believe AI can be applied is “being a founder”.
There’s just too much invested, in terms of beliefs and money into the idea that founders are special and therefore deserve seven-eight figures off of the capital pumped into their unprofitable products.
You’ll see it here in comments. People will defend A”I” applied to software engineering wherever (not) possible, but building companies? Now listen buddy there’s an irreplaceable human genius at work.
Exactly, I always find it ridiculous how the suits, any layer of mid-managment to executives, are so eager on AI 'outsourcing' everything, but they themselves think the 'outsourcing' (if it really works) would stop just before their position.
I agree with you and adding on to it,
Anecdotally, Someone I know said that their manager just asks them if they have drank water or not and motivate them, and the company wants them to make on their personal time what they are building with AI and showcase it.
So they were joking that they were building a replacement of their manager using Claude, and although they were joking, but only partially.
If anything, the managers, rationally speaking might be the one to be outsourced.
Of course, though recently I have come to realize that world isn't completely rational but I have found that on the long term, it still rewards for rational behaviour (sometimes) so I think that these managers on Linkedin should feel a bit scared... (as they are probably just prompting AI to write the linkedin-texts)
What if they are already scared and this is their way of coping with it? Nobody wants their job to be redundant and I wouldn't say that tech is completely scott free but rationally I find that tech has this taste/authenticity factor and there are too many factors but I must admit that these AI companies are succeeding in trying to force the narrative about jobs being completely redundant combined with job market, to make the labour vs capital divide even larger.
I feel as if this is why people want to be on that side of the battle and this Linkedin-Ai-founder-syndrome is a symptom not the cause.
We engineers are trying to optimize from the tech side through let's say AI to building lots of tools that we otherwise couldn't have had made due to lack of time but we had many ideas from any issues we face at tinkering/work etc..
and the managers not having to do any of it, lack the ideas in that aspect but they succeed in trying to project something even if the reality of that thing doesn't exist. So they are trying to do that.
I don't know, I don't believe much in the divide of engineer vs manager but the fight is larger than it but we are infighting...
We are all just fighting to get to the other side of the line of a broken system but oh well, so it is and I am unsure in what capacity can we fix it but I still hope deep down that things will improve for better because I guess hope is what makes us human.
Why would that be true? Successful founders have to be unsentimental by nature. If you make it harder on yourself than it has to be, you just get killed by people who don't.
You know, I've been fairly convinced we could automate CEOs away since... ChatGPT 3.5 or thereabouts?
I demo'd an automated agent platform here and showed it doing PRs, and the CEO asked me what i'd do for a living after this did all my work for me, and I said, "What do you think that you do that an agent can't do?".
While you're not wrong the problem is they own the company or the board listens to them. They have power they can use to keep existing while you don't. This is why AI is so scary to working professionals.
AI has changed the build for sure, it is a lot easier to build now, a lot easier to practice multiple copywriting ideas, do market research ... etc.
There is something that will never change for being a founder, you need to sell, and for that you need network and credibility. It was never about the building, its all about the selling. AI has not changed that.
> AI has not changed that.
But it has. AI can help you do market research, develop buyer personas, evaluate potential customers, create, analyze and enrich prospect lists, evaluate marketing channels, create ad copy, write sales scripts, think through objections and how to respond, etc.
Will it turn you into Jordan Belfort? No. Will it be 100% successful or effective? No. But can it help enough to make a difference? Sure, in enough cases.
Assumption: now everyone can do more of the above. The final line is still selling. So everyone will get to the sales part, FASTER. Triage will still happen at this stage, regardless of AI. You won’t be able to avoid this triage, regardless of how fast you get there.
I can't find a name to dig more but the "everyone will get" part is something strange to me. If everybody has the same capability increase, then what changed really ? some would even say it will increase the paradox of choice.. more offer, still the same amount of time to decide, or maybe more AI based decision to match the amount.. so less human understanding.
“everyone”. it’s there, it’s accessible, it’s “cheap”. acceleration will depend on the operator capability. if the final product will make a diff in the real world, it will ALWAYS depend on the entrepreneur, not the tools used.
By that logic, same can be said about code. So, ig everyone “gets there” equally fast, then nobody has an advantage
The difference is the destination. Me crunching code with llms vs you doing the same thing will lead both of us to different places(let’s say equally fast for the sake of the conversation). What will set us apart in the end is who will actually sell it, and that’s llm-independent.
Unfortunately it feels close to zero sum to me. I am getting absolutely drowned in AI generated personal sales pitches now. That obviously scraped my name/company online and automated the email. I feel sales becomes even more relient on trade shows and conferences and person to person interaction (Only talking about B2B stuff that I am involved in).
If anything, AI has made it more difficult and challenging because every customer and investor is drowning in AI-generated collaterals, websites, etc. The situation is dire in the academic world, where both the applicants and the reviewers now rely so heavily on AI that both publishing and financing has turned into a lottery.
I am positive this will settle down at some point, but the difference will always remain about your own abilities, not that of AI.
> If anything, AI has made it more difficult and challenging because every customer and investor is drowning in AI-generated collaterals, websites, etc.
In many markets, yes. If you're a software buyer, for example, your inbox, LinkedIn, etc. is filled with AI-generated sales outreach. And you know it's AI.
But keep in mind that there are tons of markets (think local services) where buyers aren't familiar with AI. They don't know that what they're reading was produced by AI, and they wouldn't care.
In these markets, if you use AI, you have a realistic shot at being "better" than your competition, and if you use it even a little bit more effectively, it can make a real difference.
I get it, but it is still more difficult to achieve differentiation from your less-skilled competitor in the short term, because they can simply slop their way through, at least until prospects realize that this is a bag of sh%t
> They don't know that what they're reading was produced by AI, and they wouldn't care.
It does not matter of they are drowning in it. If their inbox is filled with slop pitches, whether AI or not, they will stop reading.
The same thing happened in other businesses where it became easier than ever to put out content or product, but harder to break through because of all the saturation.
AI generated market research won't necessarily match reality.
Nor will human generated market research.
And? Spending 1000s of hours searching on Google, reading human-written market research reports, etc. won't necessarily "match reality" either.
AI is a tool. A starting point. A feedback mechanism. It's not the end all or be all.
If you're using AI for your marketing you're going to get lumped into a slop category, with plenty of other products to keep you company. Only people with AI psychosis actually believe this garbage. All LLM output has a cheap stench to it that's impossible to ignore.
There is no shortcut to hardwork, but llms somehow have people thinking that is the cases, it plays so well into people's desire to be as lazy as possible.
Outside of the tech/AI bubble, you'd be amazed at how few people can spot AI-generated content, and how few people seem to care if they think the AI-generated content speaks to their needs.
I know a small business that generates many of their leads by responding to posts on social media. They recently started using AI to create personalized comments responding to these posts instead of generic comment templates they used before. The number of leads they're generating from their social media commenting has skyrocketed.
comments yes, because people have their guard down. Its just lazy. Good for them I guess, until they lose their accounts or someone like me notices and puts them on blast.
Nobody cares. That's the thing. If you're using AI in a relatively "thoughtful" way to more efficiently provide information that's of value, the average person isn't concerning themselves with the fact the end product was produced by AI.
Facebook, Instagram, LinkedIn are flooded with AI content. A lot of it is "slop", yes, and it's kind of annoying if you know where it came from, but you're missing the big picture if you deny the reality, which is that if AI content is created in a way where it does have some value, average people appreciate the value and don't have some ideological hang-up over how it was created.
AI is just a tool. It can produce absolute rubbish, but it can also produce some decent stuff too.
> ... its all about the selling. AI has not changed that.
Nor did the web, or mobile, or any other innovation. That doesn't mean you can't build your business around an innovation.
With all due respect this reads a little deranged. To sell something to the masses you fundamentally need a product to sell. I'll agree that how you market the product can be a "product" in itself, but that only gets you so far. If it was never not about the building why waterfall vs agile why velocity why stakeholder why business analysts why meetings why board members pushing for features?
This is like when AI bros claim that AI has changed absolutely everything for their project but the first thing they do is reach for docker compose, react and postgres. Why don't you forget the bloat and have your LLM make your container, vdom differ and lightweight DB?
It's very surreal to have to point this out.
It doesn't read deranged at all to me.
To sell something to the masses you fundamentally need a product to sell. Agreed.
And you fundamentally need a way to find the customers for that product that need it at the time.
And you fundamentally need a way to interact with those customers that can persuade them to use your product.
Lots of aspects are vitally important to the overall success.
Especially as someone outside the US, building a startup on AI sounds like a bad idea. Some AI company fails to pay their bribes on time, or your country doesn't cede territory to the US president, the AI gets yoinked and you are left with Mistral or Qwen.
(Technically that also applies to MS Teams, Google and so on and not just AI)
The platform risk for people outside the US using US based AI products is enormous, as we just learned when Fable got yoinked.
Even worse if you build your processes around one ai provider like anthropic.
> Now someone with no engineering background can build production software that brings their idea to life, while a technically adept founder with little business knowledge can easily produce a go-to-market strategy, a financial model, and a highly polished pitch deck.
I had a bit of a laugh. The non-technical business experts are much more likely to achieve success than the technical experts. They can actually talk to the customer and get the customer to care. No quantity of GPUs and gas turbines can correct for a lack of personality or reputation. The technology is generally not the hard part in most businesses, despite the extreme efforts of certain technology people to make it seem so.
Does this include making annoying Linkedin posts every other day about how AI 30x'ed your engineering output and killed graphic design for real this time?
> The traditional startup growth arc assumes that the path from idea to scale is validate → raise → hire → build → raise again → grow → hire more → repeat. Now, AI has erased the expectation that each new phase in the startup lifecycle requires a bigger team, a different skill set, and a fresh funding round.
From 2015-2019 I spent the whole time saying "If I don't write the code nobody does". It was the point of saying to do anything requires a team, to build that team you need funding. It was a vicious cycle and took a long time to get enough traction to raise funding and do that... and then you end up in the MVP loop -> hire -> build -> validate -> rehire -> rebuild -> revalidate. Today all of that has changed. You don't necessarily need the team to write the code, it's for a different function entirely, maybe the original function which is the team was the orchestration engine for all the different pieces at play to make a company and product successful. The code is only half the equation. Looking forward to seeing how solo entrepreneurs leverage these tools and how teams transform using them.
Feels like the people who want to do it spend their time here arguing about it, and the people who will do it are the ones in the trenches epxerimenting and trying to make it work. The OP is just insight porn at this point.
> Founders who've never written a line of code before are shipping production applications, reaching revenue before scaling headcount
Stats please
I'm sure there are at least two in the world
This framework does not sound that much different from what people used to do but with AI agents and coding assistants, and this is not going to work unless you are lucky. LLMs cannot come up with good ideas and coding (if you believe it is solved) is no longer moat especially in the early stages.
So either go viral or go home.
Obviously personal connections, timing, market position all play role but let's be honest - this is not something that can be planned although in retrospect it may seem so. A % of the population will get all of this right many times in a roll but this is just mathematical certainty.
A serious AI-native founder should not waste time reading this brochures, they should make agentic loops instead where their agents autonomously read and produce brochures for their brochure first, product second startups.
I'm tired boss.
There is still so much old school thinking and process in this. Go through this process that has been the same for the last decade+ but now you just don't need a team of people to do it. You can just use Claude instead! Really, we are in a paradigm shift, not in a "do the same thing but with less people" shift.
> The founder's role is shifting from individual contributor to orchestrator, allowing them to focus on the work only they can do.
Founders are individual contributors? Hmmmm.
My absolute favorite quote:
Loss of objectivity
The challenge: Ask an AI tool for evidence supporting what you already believe, and it will find it. Confirmation bias now comes with a research engine.
I'm very grateful for this right now!
I just passed the idea stage and then hit the MVP stage with a prototype with a potential customer who's definitely said they would buy it!
And I'm noticing there's going to be a whole bunch of other things that I haven't thought of around actually running this thing and the timing is perfect for this playbook!
Reads like a Shopify ad. Just use their tool and you will be able to achieve financial independence.
Buying a welding machine doesn’t make you a welder
Tell that to certain CEOs that realized they can generate slop code and make their SWE teams deal with the tech debt lol
Good guide but I think the product market fit portion of a startup is so key that you need no other skills except that to make a good startup. AI won't help you with that portion, only in depth knowledge of a industry or natural product intuition will.
Who knows, maybe an AI ideated and AI created product will be the best app of 2026.
Why's everyone selling themselves as "AI-native" these days?
I feel like a lot of this advice is kind of dangerous. How do I draft a tight investor memo? I'll ask the slop machine!
It's kind of analogous to how I'm writing code right now. For simple stuff or low priority stuff I'll fire claude at it and won't look at the code if it works. But for the important stuff I'm very carefully integrated into the cycle making sure what's coming out at the end is just right. I'm carefully constructing prompt loops and validation cycles to make sure what comes out looks like what I want - because I have the knowledge and experience of what works for my specific use case. Drafting an investor memo seems like the second category of thing, you need it to be right. I don't think claude offers much of value there. What's more - if you start slopping your investors, you are going to piss them off. Unless Claude is going to say it has some special data source it's used to train on so it knows good from bad, I think this is a bad idea.
This article also kind of fits in the category of "Here's how to use AI for EVERYTHING!" and actually it would be far more valuable to say "This is the bits that AI is good at, and here's where you need to do it yourself" - which is obviously a position that Anthropic can't hold.
What's AI-native these days?
I've noticed that seemingly every single tech company has re-branded themselves as "AI" company. Add a RAG system and you're now AI. Add a AI-chatbot, and you're now AI.
not much meat about marketing
If this is true why am I hearing so much from “AI startups” about how they have big plans to hire?
Are people upvoting this so more people can laugh at it? This whole post comes off as a parody of what Anthropic would say.
step 1: find a problem people are willing to pay to make go away.
step 2: find a way to solve that problem for less money than they are willing to pay
step 3: AI???
I wish there was something similar to this, an online marketing model. Indiehackers definitely need it.
Why is this a PDF? What if I want to read this on a phone?
Mostly marketing fluff
Maybe it's better for startups to go with natural intelligence instead of artificial intelligence, just a thought
I wish hyperlinks used underlines. The worst possible UX is making your hyperlink resemble normal text, especially on a dark background. Sigh.
Here is the direct link to the slides:
https://cdn.prod.website-files.com/6889473510b50328dbb70ae6/...
They've got underlines for me at least
IMHO most founder will fail because leaning to heavy into ai and creating a system where they never experience the friction necessary to build the domain knowledge which ultimately could be the deciding factor.
Just think about website design, I don’t think it’s far-fetched to say that a non ai design website will outperform an ai designed one. These percentages add up in multiple disciplines.
I would argue betting against ai is your best chance of succeeding frankly (not in all cases but certainly as displayed here)
I think it's easy for those already in the tech industry to pooh-pooh this, as the previous comments on this post have.
Right now, people with ideas prompt their LLM by saying "I know how to make x, how do I turn that into a business?" Anthropic knows that, and releasing a playbook like this is a way to make people who haven't asked that question think to ask it.
For a non-technical person with a small business they don't know how to operationalize, an agentic workflow is a game changer. You might go from only getting 30% of your work time to build and improve your actual product to 50% or 70%.
Can you imagine having a knitting business, and suddenly being able to gauge interest for different colors with a website selector you'd have no idea how to automate? Or needing to close your shop for an upcoming holiday, and having Google and Apple Maps and your website all updated to reflect your closed dates cleanly, without having to fight through every UI? An engineer goes "bah", a baker goes "I just got to sleep two more hours".
I truly think that people in the tech industry do not understand how hard technology is for people who aren't in it.
> I truly think that people in the tech industry do not understand how hard technology is for people who aren't in it.
When it comes to AI, a lot of them don't want to understand because it threatens their livelihood.
> Can you imagine having a knitting business, and suddenly being able to gauge interest for different colors with a website selector you'd have no idea how to automate?
This does not sound like an issue small craft businesses have, but something programmer think is a thing.
I think you're thinking of the value in "creating a selector", which is not my point.
Businesses need to gauge interest and experiment with options. Knitting colors was my example of that, since I've got some personal experience helping a knitting business.
If you're a nontechnical craft business owner, the work of figuring out what questions to ask to grow your business is absolutely nonintuitive, and it's one of the things LLMs are best at. The fact that it can also modify your website or listing or what have you is an additional benefit; I chose to use that specific benefit to make the concept potentially more legible to the HN audience.
As someone who uses these tools extensively: they're extremely productive and extremely idiotic.
Detail-oriented work with lots of output that can cover up the noisy bits of thoughtless garbage? Sure, great.
Analysis-oriented work where decisions have consequences over large amounts of resources? Only an idiot would use these tools for that.
Maybe as a conversational note-taker, but anything more and you don't know what you're doing.
this is clearly targeting either vibecoders or people with extremely limited experience in software. i feel like it would be obvious to anyone who doesn't fall into those two categories that the 'AI wrapper' product category is both flooded and has a short shelf life. so many of these products are going to get wiped out once anthropic and other frontier labs stop pricing tokens below market and charge what they actually need for profitability. this is a short term cash grab at best. but hey, they gotta move tokens.
>this is clearly targeting either vibecoders or people with extremely limited experience in software
From the PDF:
>Launch energy is generated from ephemeral forces, like your founder’s friends, prospective buyers at your investor’s other portfolio companies, or a Hacker News headline that drives a spike. Unfortunately, none of these reliably predicts what happens at week six or week twelve when that initial boost has faded
It looks like it's targeted to HN users
I never understood the line of thinking so many people are trying to push. “Yeah you can build your own CRM now but the marketing and selling is the hard part.” Who said I wanted to sell it? I’m building it for myself so I can ditch SalesForce. Do I need to sell it to others for it to benefit me?
The cope here is that no one will build there own Shopify or SalesForce or Airtable because “the selling is the hard part.” They don’t need to sell and market it for those SaaS to fall.
Claude, make me rich. Make no mistakes.
Copypasta for LinkedInfluencers and VCs
Honestly at this point I’m not even sure a non-local startup is a smart move anymore
Like what do we really still need?
Most riches nowadays are created by entertainment or scams :/
It is a smart move if you want to be that mega rich scammer.
AI psychosis at it's finest.
Link to PDF
https://cdn.prod.website-files.com/6889473510b50328dbb70ae6/...
Huh, I had to search this website-files.com website and it seems to be of webflow.
Was the PDF created by webflow? a bit ironic.
Maybe Anthropic should read their own founder's playbook to create a replacement to this tool.
(Yes, I know that when involving cdn's, there's much more to the infrastructure as well but isn't that what anthropic is promising in the founder's playbook that you can be a founder which solves problems [from my understanding],
My point is that there's probably much more to it than what might've been suggested which isn't a neutral source in the first place)
Anthropic following the well trodden path of trying to present their product as an entrepreneur’s dream.
Similar to Shopify and all the make money online dropshipping slop they produce.
When I see notes like this, I wonder whether every success story can really be summarized and patternized this way. If you're building an AI based startup, what exactly would be the point of differentiation? That seems to be the difficult part
I've been at a few VC / startup events recently and I was stunned to see the number execs frothing at the mouth about finding a 1-person-ai-driven-billion-dollar-startup. This "playbook" is probably not going to help.
Someone make a guide on how to turn these VCs into marks I can scam with a bullsh*t AI native product.
“please give your funding round to help us pay for our bills”