qwertox

"I actually have a much better idea how to use AI responsibly: Simply don't use Claude to completely vibe code AND COMMIT your code."

I usually open a new chat in the ide and ask it to "generate a commit message", then i copy paste it.

This means that if my code was written 20% by me and 80% by Google Antigravity, and Claude then adds a Co-Authored-By-line, then i think it's wrong. All it did was generate a commit message.

I never had any issues with them. And they fixed my problems in a good way.

What do you use?

> "if the product is free, you are the product"

I feel like this is no longer true. You are now the product regardless of you paying for it or not.

CPU (2012)

  Model name: Intel(R) Xeon(R) CPU E3-1265L V2 @ 2.50GHz

Mainboard

  Product Name: P8Z77 WS

GPU

  05:00.0 VGA compatible controller: NVIDIA Corporation AD106 [GeForce RTX 4060 Ti 16GB] (rev a1)
  05:00.1 Audio device: NVIDIA Corporation AD106M High Definition Audio Controller (rev a1)

Memory: 32GB

This works.

Maybe he was trying to say "look, when I can look this up, it means that you can too, i won't put any effort into it, because i pay you, but if i can use AI, so can you."

Maybe that was his rationale, "to educate" by example.

Absolutely nuts to treat employees this way, sending ChatGPT chats around.

I noticed that sometimes discussing the code with a chat instance, and having it write prompts for an in-IDE-agent, then posting the result back to the instance to discuss the results and repeat this loop yields not only good results, but also makes me understand the codebase better. I let both agents know I'm proxying and take part in the conversation.

They have completely damaged my install. I no longer have the IDE. Classic Google move.

I did an in-IDE upgrade (!), when i restarted it, it did not show the editor, but the agent manager, and there is no longer a way to launch that ide.

So i installed the Antigravity IDE, which the big brains at Google decide to install to the same location as Antigravity 2, and with both being Electron apps, does not work.

Both `Antigraviry IDE.exe ` and `Antigravity.exe` launch the same agent manager.

This is then the device where Google monitors everything you do, just like on the phones. And decides what you can use and what it reports.

They could well use the data from their own developers, couldn't that be enough?

~2015, "DevShare". They wrapped open-source software downloads with opt-out adware and PUPs (potentially unwanted programs), without the original developers' consent in some cases. They took over abandoned/unmaintained projects (like GIMP for Windows, VLC, etc.) and replaced the original download with their adware-wrapped version.

Maybe that will force the companies to not be allowed to just lock you out of the account.

> Oh and he pays engineers $500,000 to spend $250,000 on Claude tokens.

What he said was that if he's paying 500,000 for an engineer, he expects that engineer to spend at least 250,000 in tokens, to get the most out of those $500,000.

Basically that it would be counterproductive to hire a top tier engineer and restrict his token usage to let's say $200 or $500 a month.

I think he's right on this.

This is very likely not true.

The use of thermal paper was surprising. Great idea, indeed.

@sama's did say: "[..] will not be used to independently direct autonomous weapons in any case where law, regulation, or Department policy requires human control". Law is what Trump decides.

> Putting a fake case report on GitHub wouldn't make it any less fake.

Much easier to review for whomever wants to review it.

"unstructured word slop markdown"

Where's the issue? In this case there's no difference between a structured markdown and a structured JSON document for an LLM.

What's wrong with a nutrient table in markdown?

How does my chatbot use the API if not through MCP?

I want my local dm shop to offer me their product info as copyable markdown, ingredient list, and other health related information. This could be a way to automate it.

> will not be used to independently direct autonomous weapons in any case where law, regulation, or Department policy requires human control

That says it all. Those laws get issued the same way the tariffs did.

Then reject any offer from the DoW until things are fair.

I wouldn't be surprised if Sam sucked up 100% to the DoW with an NDA and an obligation to lie. He and his pal Larry are absolutely in for these kind of deals. Zero moral compass.

Keep your user prefs minimal and use project memory instead: create a new project, it will only have access to your user prefs, everything else is fresh.

Yes, but just because it has version history doesn't mean it is closer to git than to Postgres. You can also do versioning in Postgres. You can even search more easily in the history.

It also won't replace Postgres, because that is also a different thing.

> The president of peace btw.

Europe is to blame, according to him.

Last time i pressed the button to delete all my chats it behaved odd afterwards, then all chats where there again. Just saying.

I was also impressed with Handy.

I played around with it this week, and when you enable advanced mode and add a post-transcription AI model to point to your own server which mimics a minimal ChatGPT-compatible behavior, then you can use it to modify the output, even return an empty string if you noticed that the transcript was more targeted to do other stuff ("turn the lights on"), if you then return an empty string, it won't inject keypresses.

So one gets the best for both worlds: transcription for dictation and transcription to trigger events.

If I now only could let it listen constantly and react to voice, so that no push to talk is active, that would be nice.

Maybe this project here could be used for that.

Also, this seems to support streaming transcription.

I would rather have them commit to make a standard out of --sdk-url. I really want to use it in production, but it being undocumented means they can take it away anytime, so stdout it has to be (and hooks).

It does list the dedicated servers price increase:

https://docs.hetzner.com/general/infrastructure-and-availabi...