Good luck to him but the 'vibe coded' angle is just his expensive PR team at work I would say.
The machine never stops! Probably it never sleeps, too.
And that’s with far more stacked teams with far better credentials e.g. Perplexity
Is there something especially good about Base44? Or are they paying for the user base or the time to market advantage?
Also the founder possibly spent a bunch of his own money it. I mean he hired 7 people already.
/s
> the company was profitable, generating $189,000 in profit in May even after covering high LLM token costs, which he also documented publicly.
Still a crazy multiplier even if we do not assume accounting shenanigans.
But it's something that you can put in your company's portfolio to attract VC capital.
It's a terrible tool that makes a lot of money for the corporation.
The real answer is that vibe coding is what people are calling it when you just give prompts to an llm and it writes the code for you. So a vibe coding platform is just an app built around that purpose. A conversation window with a a bot that writes code for you, with a bunch of integrations so that it can do things like databases and auth.
Probably would have been quicker to Google it than to ask here though to be honest :>
Examples include Bolt.new, lovable, mgx.dev, google AI studio... Vercel and Replit also have popular platforms. Google AI studio also counts, here. There are probably dozens more.
I'd guess in a platform they probably pre-prompt it with a context and rules on top of that so users don't have to say things like 'make sure AWS keys are not in the code' and 'users should be authorised for only their data' etc, and give it tools to use to make standard features.
No idea what the end result of the finished apps is though.
Right now they're something that lets all of those "I have an idea for an app" people actually make the app themselves.
In a few years, they might be much more than that.
They do not work, from my experience; or the experience of people using them: https://github.com/stackblitz/bolt.new/issues
Also $25 million of that is essentially a retention bonus for the employees.
Don't have proof, but the discord community and the WhatsApp groups tell a decent story.
See my other comment in this thread for more comments.
It means you tell the product what you want and it gives you the code.
I don't know if these 8 people vibe coded the vibe coding product itself -- they may have. But in the title the reference to "vibe coder" refers to their product.
The maintainability of the output of your average vibe coder is going to enormously exceed that of your average Wix site. LLMs are a massive threat to their whole model which relies on increasing levels of esotericism.
Companies like Squarespace set the bar much higher. Wix has always been a nightmare to use and customize.
I'm not implying anything! Wix is clearly actively designed to be that way.
When I was evaluating user friendly web platforms for several non-profits, I ended up choosing Wix over Squarespace because Squarespace didn't provide backup>restore functionality. That's a pretty low bar...
Many of the best (and successful) restaurants near me don't have websites, for example.
In some places people run their businesses almost entirely through chat apps like whatsapp or wechat. Works great for businesses based around client relationships.
However, any self respecting business owner planning for a stable long-term presence should absolutely also have their own website, and have it made to be as independent, simple and easy to maintain as possible in the context of its technical needs.
I went with Wix, as it was the best of the builders, but it is still totally crap. I do not understand why web tech has moved so slowly in this area.
My dad has a Shopify site. He'll sometimes accidentally break his theme by making simple modifications in the editor, and it's not obvious why. Maybe a missing quotation mark or a wrongly nested closing tag. He usually had to revert it or even ask for help.
99% of the things he wants to change about his Shopify site would be better done by asking an LLM to change the template for him.
Frankly, I don't want to be editing HTML by hand to make simple edits, either. I just get Cursor or Claude to do it. If I were using a platform like Shopify or Wix where I don't have the templates sitting in my filesystem, I'd like an LLM integration there too.
Almost all of us have an LLM integrated into our code editor, yet when people want to integrate an LLM into the workflows of non-software-devs, suddenly there's cynicism.
Funny - with a non-profit I help out I was so frustrated with our Squarespace site that I ended up replacing it with a custom site I vibe-knocked out in less than a day.
Better design, faster performance, integrated into mailchimp and volunteer backend, better donation flow also integrated, custom page type templates, better cms (sanity), hosted for free now etc. etc.
Having a semi-decent developer crack open Cursor / Claude Code and spend a day just building _really_ is a threat to all of the hosted website and CMS providers. I'd definitely include Wordpress in that too.
The problems begin when you try to do things slightly different, and this is almost entirely down to the staggering levels of inconsistency in their core APIs. (Genuinely impressively so). Their business model relies on the useful pieces being opaque enough and irregular enough that you cannot move off without overcoming a big migration effort.
Consequently their interest will be in two areas: can they use LLMs to get people to use those existing modules and get locked in faster? Can they own the funnel of people using LLMs to attempt to create this sort of functionality for their existing market? This acquisition helps on both levels, so it makes perfect sense for them.
I am not a paid member of the team, just an admirer who wanted to get closer to the action. This felt right to me from the first moment, and I'm happy I had a small part in the journey.
I met Maor (the founder of base44) and team and had beers with them. Good people.
--
Let me clarify a few things:
1. I don't know exactly how much vibe coding went into building base44 itself. I can attest that Maor's rate of releasing features was absolutely insane - I'm talking major updates every 1-2 days. I assume he's good with Cursor and the like. He's also very, very decisive on what to build and what not to build. Aggressive, even, I would say.
2. Maor had, for the majority of the life of this, no team. The employees joined way after base had customers. Most of what Base is was built by Maor, with 1-2 close friends helping cut out everything that wasn't relevant or wasn't great (so I'm told).
3. It's a different take on lovable/bolt etc. No one argues this.
4. Maor opted to include the db within the platform, rather than enable persistence externally. This really made the output great, and made fixing cross-application things very easy.
5. To me, base44 is PHP. It's a bit ugly, but it works, easy to explain to people, and once you get a hang of it it's a great hammer. It's not going to win the space race anytime soon, but it'll build you a house.
6. Base has resolve with AI functionality, which is far superior to anything I've seen outside of an IDE. It just works.
--
To folks trying to win the AI race by building exceptional technology on the bleeding edge, good on you. I don't think Base is that.
I think Maor symbolizes something different: we're in the fast-grab era.
Big cos are not able to build killer AI apps at the rate they're expected to, which means they're circling around looking for what they can snatch with money/equity.
My take?
Build AI things that just work for a specific use case. Release them fast. Make people fall in love with them because they "do AI" for the use case.
Some bigco is flying close by, trying to build it but failing. Be there for the purchase.
They weren't there, and then they were there.
Looks about right. see 6
User Enters Prompt > Combine with Internal Prompt > Feed to AI > Return Results.
That's pretty much it but realize the "internal prompt" will take you months of work and refinement. There are many many tiny edge cases around using AI like this. One problem for example is you can tell the AI to return you it's data in a standard JSON object that you app can use, but if you set the temperature to high then the output will stop conforming. So you start having to do "pass-offs" between conversations with different temperatures.
And this is just one of a litany of issues around using AI like this. And really this is no different than a "traditional" SaaS app that doesn't use AI, I have a side project and that project has a litany of issues that must also be fixed.
Bottom line is building something like this is not any easier than building a traditional SaaS, just different flavors of problems.
The other day I tried to legit vibe code for about an hour and Claude could barely build a weather app that used NWS data for me. Even with this toy weather app it collapsed a few features in, simply put the context windows of these AI tools is just not large enough. I think vibe coding _might_ be viable once context windows start growing into the 5-10M token range but even then I'm skeptical.
A pithy statement that is like candy to VC ears, but falls down if you think about it even a little.
Nobody is going to buy Adobe for its client list.
How much would the code of say Photoshop be worth if the entire Photoshop team were to leave and set up a competing business?
An early stage company that barely has any code? Yes, in that case, it's the team that matters because they haven't fulfilled the vision yet.
This is strikingly similar to just about every tech acquisition in history. Buyers purchase the assets, potentially including part of the team, needed to produce a reproducible business outcome, however one defines those terms.
What is striking is how quickly the team was able to assemble a reproducible business outcome worth so much.
Maybe with a little bit of vibe coding they can fix their infrastructure. But I doubt it.
"Solo unicorn.. well, ok it was $80M not $1B.. and, sure ok it wasn't 1 guy, he had a team of 8.." LOL.
Guy lists himself as an "angel investor" and "Forbes 30 under 30" on his Linkedin, so clearly not his first rodeo and has good connections, likely a stronger ingredients than the tech.
So you know, not bad money, and I'd certainly take it. But doesn't seem like any step function change in ROI for startup founders due to LLMs here.
As others have pointed out, Wix bought the leads/client list.. not the tech.
The article mentions that $25M is a retention bonus for those eight employees. (About $3M each — a nice bonus by any standard, but not insane for SV + AI. Presumably that's Wix stock, not cash.)
So the purchase price might be made up of components something like this:
- $30M for the client list
- $25M retention bonus to employees, with standard RSU vesting
- $25M for the founder, with earn-out goals over four years
Neither of these are signals :) It looks like there are much stronger different signals out there for this guy.
I quickly built a retirement calculator, two prompts and deploy. https://app--future-focus-dd96b32f.base44.app/
Basically everyone in the israeli tech community know about this guy because he pretty much shared everything about his journey from the moment he started. One of the most genuine and no-bs entrepreneurs i've ever seen (and israel has it's fair share of grifters).
Yes, he really did basically built everything by himself. Yes, he did manage to build a huge client list worth a lot of money, in just 6 months. Yes, he was previously a founder (without an exit). Yes, there were a few (very recently hired) employees.
This is one of the rarest solo-founder exit stories in the world. dumb-founded by negativity.
To be fair, this site also has lots of wonderful posts and many excellent or at least thought-provoking comments, which is why I still visit frequently.
Is this a vibe coded product
Or a product that helps you vibe code.
Or both?
They can’t have 8 of them vibe coding. And if it isn’t vibe coded then this has FA to do with a one person AI unicorn.
Edit: Just learnt from comments, they used Claude and not ChatGpt.
theHolyTrynity•7mo ago
shawabawa3•7mo ago
noosphr•7mo ago
Q: Is it true that Ivan Ivanovich Ivanov from Moscow won a car in a lottery?
glimshe•7mo ago
Snuggly73•7mo ago
In the newspapers there were news that Bulgaria sent 8000 personal computers to Japan. On the next day, there was a slight correction published:
1. It wasn’t 8000. But 8000000 2. It wasn’t computers, but jars of marmalade 3. They weren’t sent to Japan, but returned back by Japan
harvey9•7mo ago
I think anyone who was ever measured on LOC and Jira tickets can relate.
uncircle•7mo ago
also
https://en.wikipedia.org/wiki/Russian_political_jokes has some great ones
trhway•7mo ago
simion314•7mo ago
bmacho•7mo ago
moralestapia•7mo ago
- HN rules.
hobolobo•7mo ago
https://www.bbc.co.uk/programmes/p03mndbk
mellosouls•7mo ago
https://www.youtube.com/watch?v=mN3z3eSVG7A
hasbot•7mo ago
jan_Sate•7mo ago