I think “software factory” just means automations where bots pick up tickets and spin up code changes. Maybe Stripe has that many well defined bugs but there’s no way it works if you’re building anything remotely complex. There’s just a lot of nuance that agents cannot pick up
This is just a bet that the models will get better and that all the half baked never seen production Lovable apps will be able to be brought life because of that. I am fading that bet personally and am bearish on non technical people building software
If the models get that good you won’t need Lovable at all. That’s the strange part - the model companies will instead eat the infra layer (at least I’d imagine that’s what happens - Claude just knows how to manage AWS for you). This seems like a very shallow moat into which to pour $400M.
People still need a platform to host their app. AWS is far too complicated for the average person. Even with AI to help. The user still needs to get past the registration screen and put their credit card. Most will panic and run as soon as they see the letters "AWS", "IAM" and "EC2"
Then as soon as their agent hits the first bug "port already in use" they will run... Any brave soul who remains, after spending a whole day solving similar issues back to back, will be like "The port has been freed up, my nginx proxy has been configured, my nginx has been launched with systemd and my app server too! The socket ulimit has been increased, I still can't see my app!!!" Then Claude be like "Well of course, silly human! To achieve this simple task, you would have to update your security group inbound firewall rules, duh!"
Then the user be like "Claude, we fixed everything! I cannot see my app still!" Then Claude be like "You goofball, we didn't even get round to purchasing your domain name, moving it to CloudFlare DNS and configuring the A record, CNAME and AAAA record for IPv6 in your Zone file... Not to mention the MX record. You also need to obtain a signed SSL/TLS certificate from Let's Encrypt, then modify your nginx config to install the SSL certificate... Oh then we'll need to configure your nginx to act as a terminating reverse proxy for your WebSocket connection... Oh actually before we do any of that we should set up your schema validation logic and access control logic on your app server with rate limiting and backpressure monitoring... Then we'll set you up on Git version control with a CI/CD pipeline with precommit and PR hooks to run your test suite and code linting... Then we want to keep the registration and login flow of your app real simple, like stupid simple, so we'll get you a Keycloak instance acting as an OAuth2/OIDC identity provider for authentication with an email verification step which will involve us configuring the realm and creating an OAuth client and setting up the SMTP email server... Too easy!
Totally agree - but do you think that is a 13 billion dollar moat that Anthropic would not be able to solve through branded integrations and connectors?
I’d imagine they’re already working with folks like Stripe, AWS, etc on exactly this at the moment potentially? I guess that’s where Lovable comes in. And now the money makes sense. Do the raise and become an acquisition target so one of the labs doesn’t have to do it themselves.
I suppose that Anthropic or any of the other companies that develop foundational AI models do this. This would be a vertical integration strategy. However, they would spend less time developing the best foundational models, so they might not want to keep a more focused product vision, and keep their capital focused on build better models and improving Claude.
Yes, they will be able to do this, but it's a lot of work, and Lovable has a pretty good head start.
Frontier model providers are already doing this. Claude has artifacts which are served on an online URL, and OpenAI has sites, similarly. I don't think it's a distraction from their model making, those are separate teams, and thus we don't need Lovable, the frontier model providers will vertically integrate.
They’re good enough already. You can one-shot a website with Claude Code or Codex of the same or better quality than Lovable. I have no idea what the appeal of the product is at this point.
may help thinking through what you'd have to do to go from the generated web app to get it running on a website, may be with subscribers, payments, etc. and then operating it. There are so many concepts, constructs, terminology one needs to know to do it without ongoing headaches. Lovable at least takes away some of those pains.
> at least I’d imagine that’s what happens - Claude just knows how to manage AWS for you
AWS has $100+bn reasons per year to not let that happen. As techies learned slowly through the evolution of the internet: political power > money > tech.
Because they don't threaten their platform, they're a customer. If Vercel ever becomes more than a flea and somehow realistically threatens AWS' success, they will either be acquired or hobbled.
Im bullish on more people being able to build software than pre-AI but bearish on the general population being able to. The unfortunate truth of AI is that it requires you to think like a dev to be effective. And thats just not how a lot of people think or want to think. Thats fine. But that means AI wont be a huge unlock for the masses.
I think we might underestimate the evolution of these tools as AI gets both better and cheaper. The current form factors are like trying to setup a website in early 90s, cumbersome and not user friendly. Coding infra, AI coding UI, etc are not solved problems yet. If things progress, I think there will be a time when a whole app can be developed and deployed with a single prompt. In that kind of world, who can’t write code?
I mean what Im hearing is "once we have AGI it will be easy". Sure, agreed. But AGI is going to unlock a lot more than software dev including having people just one-shot Lovable replacements. I also just fundamentally disagree its going to happen without another several "attention is all you need" events all chained together.
I think you are doing the equivalent of looking at a hockey stick graph and extrapolating to infinity on where we are headed. We are already hitting the asymptote and the line is starting to go horizontal. Theres plenty of gains to be had by current state of AI, I'm not trying to be a wet blanket. I just dont think those gains are going to be had by "the masses using AI". Its going to be highly technical people doing highly technical things at scopes and speeds previously unimaginable.
I think you're failing to see the forest through all those trees. The value of software will plummet as barriers to entry fall and people become more capable of creating exactly what they need for their own specific purpose.
>I... am bearish on non technical people building software.
I can't bring myself to be bullish or bearish in such broad strokes.
Over the years, I have met and worked with a number of people who were "non-technical", but who nonetheless built their own internal rough software tools with a cobbled-together combination of spreadsheets and SaaS services, connected by something like Zapier. They are a particular kind of person, but I would not call all of them them "technical". They are experienced in their business domain, are detail-oriented, and are persistent in making improvements by any means they could.
I have seen those people be supercharged by AI app builders like Replit and Lovable, even if I don't personally have a need for either.
This might be a matter of definitions: Is such a person "technical" because they value systematizing human processes and put effort into it? Or is such a person "non-technical" because they are not experienced in software engineering best practices, or programming in the conventional sense?
In addition to the nature of the user, I suspect the fulcrum that success and failure rests on for these kinds of apps and tools is long-term maintenance. How frequent? How involved? How breaking are the changes? On this topic, there is a quote I think about a lot (paraphrased, and I forget who said it, or I would credit them):
>AI app builders mean that anyone can experience the joy of building software, but also that they will experience the joy of maintaining software.
I have seen and worked with a number of “non technical” people ship apps and automations over the last year that are making real money so it’s past the point where you can be bearish. Sure these people are probably just a class in themselves and not “the average person” but they never would have done this if they had to write code by hand
I mean 200k/MRR for one company most of these are not pure tech but people in a domain (the 200k one, while I don’t feel comfortable outright stating it is a car buying application you could likely find with a cursory search) who use tech to scale. So I have worked with lawyers, therapists, people from other countries etc to bring apps to that market using their domain/market knowledge as an edge.
Doesn't matter how many Lovable clones there are. Everyone knows the name Lovable.
This caters to the overwhelming majority who cannot use tools like Claude Code or that prefer an all in one solution.
Open source models only help Lovable improve its own margins. They can stop paying the Anthropic or OpenAI tax and tailor the models to exactly their use case and UX.
It's a sign of how much distribution matters to VCs. Whether that is enough in the medium- and long-term is unclear. I'm a bit skeptical but not very confident either way... which is part of the reason I am not a VC!
I think it's time we start questioning more of the valley's mindset when it comes to good business and software practices. From lived experience it typically results in unhealthy work environments at best and society destroying at worse. Is this really the best way to write software? I'd but my life's savings no, it's not. No way in hell.
Agree with this fully, I'm sure a lot of software engineers do too. Yet software is dominated by the valley. It's funny how easily money and greed manipulate the narrative while the majority suffer.
I was open minded about it until I did a job search for "Lovable" on the biggest career website in my country and couldn't find a single job related to it in the entire country. For any other tech stack, any major town or city would give you many pages of results. Surprising.
To be fair, I doubt a job role would include the name of a specific product. A business posts a hiring ad for a Graphic Artist, not for an Adobe Photoshop Artist.
if you've tried to legitimately productionize LLMs to deliver a specific repeatable outcome, rather than human editable output(text, code), you know the models still aren't there yet and the whole product experience feels hacky
> I feel like that was the epitome of AI assisted but quality code
I totally agree with you on this! The engineer is still actively engaging while getting the performance gains of not having to type out functions. The Engineer is still in charge vs agentic "engineering" a model + harness can spit out whatever and the engineer is left to review tons and tons of code
Everyone is building these same looking apps that look like Codex, Claude desktop. We don't need these. Slackbots work more than well enough to collaborate with agents.
just a data point, at my company we are building this internally. if you're targeting people building from scratch it might work, but there's no way you can port any somewhat mature infra stack, nor will the org want to. you'll need to deal with the variable complexity of everyone's dev environment which already doesn't work locally for thousands of different reasons.
Agreed that at the moment it's a very difficult problem, but one we're looking to solve! I think it becomes a no-brainer for most people if we're able to give each agent a replica of their production stack.
What does your current setup look like? And are you using an open source solution like OpenInspect for your in-house version, or building it from the ground up?
Building from ground up using OpenAI Agents SDK. We already have custom in house Cloud Development Environment, so the effort is just to "agent-ize" though which is not small. That's why I feel any team with any sort of infra support likely won't buy your product, since they already have the tribal knowledge to set this up. Newer teams might. But overall porting people's dev envs into the cloud is a tarpit problem(IMO), I was interested in this space too but decided against it for that reason. Happy to be proven wrong though!
I think for use cases like that, we'd offer on-prem deployments (similar to Factory), potentially coupled with a FDE. Still need to do a lot more research into the enterprise space.
No one is going to churn Slack for this just because it is "Agent First". Slack is more than good enough to tag Agents and will fill in the missing pieces slowly
Slack's approach is NOT good enough, because the version of the Agent that lives in Slack cannot see the context of the Agent that lives inside Cursor on my developer laptop. They might be clones with identical brains, but they can't talk to each other or compare notes.
So why can't the cursor instance on your laptop have a slack bot? I've had a slack bot on my build machine written in chicken-scheme of all things for years, I feel like the great people (or clankers?) at cursor ought to be able to figure this one out
It could! An MCP would actually be great but Slack would prefer to sell us their own AI integration product so (1) they make it difficult and more importantly (2) it's against their terms of service
We pay them a LOT of money every year and they would absolutely notice the indexing at the level we'd need to do to make this work ourselves, so the answer is "because Slack is preventing us".
yes but we do alot more that may not be clear at first glance, things like brand data, and for scraping handling pdf, ocr, docx, ppt, xlsx automatically
shipping a bunch of new things soon which should make it clearer, but as of today yeah
Large model companies will likely build this and make it better. It'll also be cheaper overall since they'll be subsidizing token cost if you use them directly vs third party router paying API costs
I would argue they do not have a good incentive to build this and make it better. Why would Anthropic route Claude Code traffic to DeepSeek (at 20% of the cost)?
reply