I got you an my watch list for quite some time and will definitely give you a try.
It sounds promising, because it is on the outside reproducible deterministic component generation in a modern fashion as far as I understood it.
I build a large platform using a methodically comparable approach I suppose, albeit in the pre-AI time, and that's why I wanna have a closer look at the inner workings and results of your project - curiosity so to say.
You appear to be the only solid and promising endeavor in the GenUI domain with a solid approach other than simply relying on an LLM but using math in combination with AI.
I have a recent background as a member of the Senior Management of one of the largest banks in Europe (8 years stint).
I worked in IT - and build the only successful platform for financial services done in-house and to ever make it as a standard not because management tried to force people, no because it solved customer's problem (external and internal customers aka employees).
I don't buy any of these EU is going fully independently. EU's IT isn't capable of doing this. No way ever ever.
I will 100% of the time bet against it. I saw so many things, experienced so many things - and no way is there anyone out there in Europe who to this day simply can acknowledge and appreciate the marvelous work and evolution that for example Google took or any other startup to leading global business like Apple, Facebook, Microsoft, Amazon and so on.
Even IBM's mainframe: if you do not understand the very problem they solve - don't talk about "independence from X".
There is no successful startup from Europe, that has any results like the mentioned. All have to do with IT, all went from idea to what they are now.
And Europe now wants to do what?
All my fellow European's here boasting around: I feel sorry for you. All the "Let's build our own Google" (Google search) folks that predated the other independence stories like this here simply disqualify themselves the moment the say out loud such sentences.
Read 5 books about Engineering at Google, read the HTML5 spec by searching for Ian Hickson, go to the Computer History Museum in Mountain View, CA - and tell me that EU is going well in IT independence in taking a totally different approach, ignoring every and any circumstance and context the mentioned companies had, using design by committee, state dictated orders, establishing standards by enforcing them by law instead of evolution and being the best there is.
Good luck, we will see how it went.
And no, building something that looks somewhat okish from the outside, but is utterly crap inside - we talk about, excuse me my corporate language I was penetrated with, "best in class world leading number one" apps that leave any comparable solutions in the dust - just as Google, Apple, Microsoft etc. did.
If you cannot win in free markets - you declare victory by misusing regulation. That's cheap and well, socialism.
All of you will beg that all the systems and providers will still do business with you after years of trashing them.
Are Mastercard etc. awesome? Are they objectionable? Of course - but trashing them, boasting you will easily beat them while having the double standard on relying on them freely is disgusting.
All EU patriots: Throw away your iPhones, deinstall all the US apps - use at least existing Open Source alternatives. Cancel your Netflix, Microsoft Office subscription - do it.
Good luck anyway. After all, you are doing a live experiment on the head of the people forced to live with a minorities decision.
I disagree that this submission has no value at all or that it should be auto-banned — there is a difference between thoughtless vibe-coding and spec-driven coding, guarded by tests. It seems from the next thread that this project is hopelessly outmatched by the sheer complexity of taking on something as the Docx spec, but this project has value, and for someone, it may be all that they need.
However, I do agree with your point about live debugging. In light of that, I prefer to treat this submission as a curiosity about current model capabilities, and let the authors keep improving this project if they find it worth their time.
Let's be more respectful to the differing goals of people
Seconded. I could ask the LLM myself and see what it comes up with in 5 minutes, not to mention that all of this was done 1000 times already so I have no interest of doing so.
Also, it's effortless. Not interesting at all if you can't share any insight about the project, because you don't know how it works under the hood and how many architectural problems were solved (or not)
The real tragedy is that I can only give a single upvote. This post needs to be added to the HN guidelines, I feel.
We are being overwhelmed with slop. Not even original slop, but carefully specced slop that poorly replicates some existing functionality, but as a SaaS.
This is the 90s equivalent of "Doing $FOO, but on the internet", only it's "Copying $FOO, but with added costs as a SaaS".
That it is AI is just another black mark against it.
AI is taking over Senior Devs' Work is the same as IKEA is taking over carpenter's moat - no, no, and again no way.
AI lets you do some impressive stuff, I really enjoy using it. No doubt about that.
But app development, the full Software Delivery Life Cycle - boy, is AI bad. And I mean in a very extreme way.
I talked to a carpenter yesterday about IKEA. He said, people call him to integrate their IKEA stuff, especially the expensive stuff.
And AI is the same.
Configuration Handling: Works on my machine, impressive SaaS app, fast, cool, PostgreSQL etc.
And then there is the final moment: Docker, Live Server - and BOOM! deployment failed.
If you ever happen to debug and fix certain infrastructure and therefore deployment fails - you wish you were doing COBOL or x86/M68000 assembly code like it is 1987 all over again - if you happen to be a very seasoned Senior Dev with a lot of war stories to share.
If you are some vibe coder or consulting MBA - good luck.
AI fails so bad at doing certain things consistently well - and it costs company dearly.
Firing up a Landing Page in React using some Tailwind + ShadCN UI - oh well...
Software Design, Solution Architecture - the hard things are getting harder, not cheaper.
IKEA is great - for certain use cases. It made carpenter's work only more valuable. They thrived because of IKEA, they didn't suffer. In fact, there is more work for them to do. Is their business still hard, of course, but difficult in a different way (talent).
And all doomer's talking about the dev apocalypse - if AI takes over software development, who is in trouble then? Computer Science, software development? Or any and every job market out there?
Think twice. Develop and deploy ten considerably complex SaaS apps using AI and tell me how it went.
Access to information got cheaper. A fool with a tool is still a fool.
I think Vercel mixes skills and context configuration up. So the whole evaluation is totally misleading because it tests for two completely different use cases.
To sum it up: Vercel should us both files, agents.md is combination with skills. Both functions have two totally different purposes.
I rose from Dev to Senior Management in a 100k+ global banking enterprise.
I don't want to trash anyone. Having said that, I always kept my engineering approach as opposed to being a manager in the sense that what I did became an end in itself.
I was more of a renegade within corporate, and used this unique position to achieve fun and results way above everyone else. I got proof, this ain't no bragging. It was easy mode, I used the top notch devs I could hire and automated everything, build a platform, that became internally the de facto standard, which caused 600+ Mio EUR cost savings within 4 years and counting with a headcount of 8.
Long story short: I was a bit Googly, knowing them a bit and having been there.
Here is the gist: To this day I could never grasp what my manager collegues or their peers and directs were doing. I asked many and many times of any rank, because I wanted to learn.
Most things were related to administrativ stuff like vacations permissions, performance reviews, budget "planing" - and of course meetings, meetings, meetings.
95% of what the HIPPOS with high 6figure and 7figure incomes in the room were doing could easily been done by an intern, except for the people affais.
Only requirement is discipline to sometimes just sit still in a chair and jumping via Zoom from meeting to meeting every 30 minutes from 8:00/9:00 to 19:00. Monday to Thursday.
All you have to do is rely on these phrases: "What are the next steps?", "I will delegate this to...", "Now start the reports please."
These people were IT managers - of course no one except me had any (!) Computer Science background.
Google taught me, that it is totally easy to train a computer scientist business skills, but impossible to train any non-IT person Computer Science. This holds true.
So yes, I can totally relate to these news here, however I feel sorry for the people anyway. Good faith in most cases has to be used. That they do everything to appear irreplaceable and therefore cause havoc along their "career" is only the flipside of human behavior and dysfunctional settings.
Take care of your craft and be proud, if you are in need.
The class system is embedded within corporate structures. These management roles are simply a way for those higher up the class structure to extract money from the productive workers. You'll notice when nepotism happens and senior leaders kids or associates are hired it is usually in a management capacity. It's very rare to see them hired in an IC role. For example look at Jensen Huang's children at NVIDIA, and their children who have done internships....
It’s also strange because I highly doubt Google has manufactured a billion physical units of anything. Most of their consumer hardware is designed and built by partners, including Pixel.
Self hosting sounds so simple, but if you consider all the critical factors involved, in becomes a full time job. You own your server. In every regard.
And security is only one crucial aspect. How spam filters react to your IP is another story.
In the end I cherrish the dream but rely on third party server providers.
The legendary Jim Butterfield needs to be remembered in connection with the KIM.
He wrote the “The First Book of KIM” and it kickstarted his career within the 6502/6510 microprocessor family namely Commodore.
He is such an awesome role model to this day in explaining complex concepts to the average people that made them hungry for more.
Search him on YouTube, you will want to start BASIC on C64 the moment you watch him unpacking a C64 and plugin it in to show how easy it is to write BASIC programs for fun.
It sounds promising, because it is on the outside reproducible deterministic component generation in a modern fashion as far as I understood it.
I build a large platform using a methodically comparable approach I suppose, albeit in the pre-AI time, and that's why I wanna have a closer look at the inner workings and results of your project - curiosity so to say.
You appear to be the only solid and promising endeavor in the GenUI domain with a solid approach other than simply relying on an LLM but using math in combination with AI.
Good luck!