This may be the largest AI-generated codebase right now, by a lot. It'll be interesting to see how this plays out.
Frontier AI software development still falls short in the design/architecture department, in my recent experience. Though it's pretty impressive at making "working" code.
This being a fairly direct conversion from one language to another, even keeping the same interfaces across files, means the architecture is already in place.
The detailed test coverage is also very helpful for Claude. But even detailed testing can't cover every edge case.
So my questions are:
How well did Claude do on the edge cases?
And how maintainable will this codebase be going forward?
> This may be the largest AI-generated codebase right now, by a lot.
I'm sure there's lots of other large scale applications of AI, just not many/any projects that are open source and so high profile - with the changes being done so far.
Personally, in the past 3 months I've shipped about 2.3M lines of a legacy project migration, though the new codebase is Java + Oracle ADF because of reasons™ and instead of being an interesting codebase, it's more forms heavy and essentially acts as a front end for a large Oracle instance, think more CRUD than application runtime (with an upsetting amount of XML).
The difference also is that it wasn't migrated by using AI on every file, but rather dumped the DB schema into JSON, and converted the old form contents to a YAML intermediate format that describes what's in the forms and have been iterating ever since of creating code that generates code - basically AI assisted development of a codegen solution + AI assisted sidecars that get merged with the generated code based on markers, when something can't be automated that way and often times also AI controlled browser based testing (since Playwright is in the cards for everything, but not yet).
Seems to be going pretty okay so far, will probably take months more of iteration and fixes, currently the automated testing is taking a while because let me tell you - not only Oracle ADF is shit, but so is WebLogic, like fuck I'd be so closer to being done if I was allowed to pick Python + HTMX or even Java + Thymeleaf. That's still better than a team spending a year on the migration and getting like 10% of the way there.
Obviously there's no more details to publicly share, but the overall vibe is clear: as long as you can test any changes, you can iterate faster than without AI - and the code ends up being more readable that colleagues would often write. The problem is that people would squint at the suggestion of 100% test coverage previously so most code is even written in a way that is straight up not testable (and often nothing is decoupled from the framework properly and tests take way too long, both time and resources).
Couldn't this be done on proprietary software as well? Have an agent fuzz an interface (any type) for every bit of functionality and document it. Then have it build based on the document?
Very nice. I've been trying to find an image duplicate detection algorithm/system that suits my use-case for a while. Your app seems promising, however, I'm not willing to pay $99 just to see if it works with my (uniquely challenging) duplicate images.
After realizing there was no demo I was looking for a way to contact you directly with a few sample images, but can't find contact information on the website.
Consider adding a demo and contact info.
Otherwise, the app is looking solid. This seems like a great use of AI.
What are the Top 10 Elements of the Authoritarian Playbook?
1. Divide and rule: Foment mistrust and fear in the population.
2. Spread lies and conspiracies: Undermine the public’s belief in truth.
3. Destroy checks and balances: Quietly use legal or pseudo-legal rationales to gut institutions, weaken opposition, and/or declare national emergencies to seize unconstitutional powers.
4. Demonize opponents and independent media: Undermine the public’s trust in those actors and institutions that hold the state accountable.
5. Undermine civil and political rights for the unaligned: Actively suppress free speech, the right to assembly and protest and the rights of women and minority groups.
6. Blame minorities, immigrants, and “outsiders” for a country’s problems: Exploit national humiliation while promising to restore national glory.
7. Reward loyalists and punish defectors: Make in-group members fearful to voice dissension.
8. Encourage or condone violence to advance political goals: Dehumanize opposition and/or out-groups to justify violence against them.
9. Organize mass rallies to keep supporters mobilized against made-up threats: Use fearmongering and hate speech to consolidate in-group identity and solidarity.
10. Make people feel like they are powerless to change things: Solutions will only come from the top.
This feels like a decent list. I'm not an American but some of these processes seem to be happening in other places.
1. Is all of us, on the "right" or the "left". Let's not do this.
2. Here you could say maybe the government is doing a little. But I would still say most of the lies and conspiracies that are reverberating in our society are not originating from there. This is like 95% on all of us (or social media). 5% you can maybe blame Trump.
3. I don't really see this happening yet.
4. I would say the "left" has been demonizing the right very effectively. But sure, goes both ways. This just seems to be standard for political debate today (it's the end of the world if those guys get power). I think it's mostly up to us to push back against this. So if you're a democrat push back against casting Trump as a dictator (I don't think he is) and if you're a republican push back against all this "stop the steal" and "lock her up" whatever nonsense.
5. Not happening IMO.
6. I guess Trump is blaming illegal immigrants for the rise in crime. I don't think is is a perfect match to the intention here. America is so multi-cultural/diverse anyways so this tactic doesn't really work.
7. Trump sort of does this but not really to the extent that I think the author of the list meant. So far it seems there's no fear from voicing dissent. Musk went ballistic on Navarro calling him a moron and is critical of Trumps tariffs. Many other republicans are critical. This is more of a kindergarden than authoritarianism.
8. Not happening. Would be very worrying if we get there.
9. Not happening. We had large rallies before the election but you don't see the sort of things you might see in Iran or Turkey. Again this would be a worrying sign if we get here.
10. Also not happening. You see universities fighting back against Trump. you see courts. you see states. you see people. If anything it seems people feel like they have a lot of power.
You seriously don't believe that pardoning people like Enrique Tarrio for violent crimes perpetrated openly in pursuit of political goals doesn't encourage violence?
I've had to read up on him since I'm not that familiar with this topic.
I guess at some level? But in comparison with actual authoritarian regimes/societies this seems to be in the noise.
> Encourage or condone violence to advance political goals: Dehumanize opposition and/or out-groups to justify violence against them.
Again, I don't think we're seeing this happen. Has Trump given some extreme element a sense that they can get away with things they couldn't previously? Sure. That was also the case in his first presidency. Is this a society shaping phenomena. Not really yet. Could we be in a long term change that will end up with a non-democratic US? Anything is possible. Everyone needs to uphold democratic values.
Frontier AI software development still falls short in the design/architecture department, in my recent experience. Though it's pretty impressive at making "working" code.
This being a fairly direct conversion from one language to another, even keeping the same interfaces across files, means the architecture is already in place.
The detailed test coverage is also very helpful for Claude. But even detailed testing can't cover every edge case.
So my questions are: How well did Claude do on the edge cases? And how maintainable will this codebase be going forward?