> (the full impact of today's frontier models has yet to be felt by the general public - to say nothing of models that will be released in the next few years)
We definitely saw some kind of non-linear step function jump in quality around the beginning of the year - it's hard to express how good Claude opus/sonnet 4.6 is now. However, I wonder if we're going to see the same kind of improvement from here? It's kind of like we got to the 80% point but the next 20% is going to be a lot harder/take longer than that first 80% (pareto principle). Also, as more and more code out there is AI generated it's going to be like the snake eating it's own tail. Training models on AI generated code doesn't seem like it will lead to improvements.
> That’s a red line he will only cross out of stupidity.
Do not underestimate Hegseth's stupidity. He's completely unqualified for the job he has and is way out of his depth. Ditto for many others in this administration.
That got me to googling around since my grandfather was born in Germany and came to the US when he was 5 (circa 1920). But from what I'm finding it sounds like when he became a US citizen that tie to German citizenship was broken. Also, prior to 1975 the citizenship only passed down through the father - it was my maternal grandfather so it wouldn't pass down, apparently. Well, it was fun to think about the possibilities for a few minutes, anyway.
Your case sounds complicated so I'm not sure, but two things to note:
1. US is one of only a few countries where children emigrating with parents don't officially declare intent to immigrate, they do it automatically with their parents. This means that your grandfather (whether he was aware or not) was still German, since German law says you only give it up if you "take action to immigrate" or something like that. Likewise every child since then (your mother and you) were born as US citizens "involuntarily" (as in you didn't choose) so you also retained your citizenship.
2. In 2021 Section 5 of the StAG law was updated to say that people born to German mothers between 1949-1975 are now eligible, it was updated since male only was seen as discriminatory. So theoretically say grandfather -> mother (born to male) -> you (post 1949). Not an expert so double check this.
Im not an expert but my understanding of your case would be that you are not even needing to apply for status, you are literally German now, and just need to request a passport (check this with the resources on Reddit I mention below).
I'd recommend checking Reddit "German Citizenship by Descent" resources. There's a couple profile names you will see there really frequently who are German citizens who can help you in finding paperwork from German government resources if needed (old birth certificates, etc.) for a small fee.
If you're a doctor or nurse Canada is definitely saying "sure, come on in!" - they're actively recruiting in the US for healthcare workers. But that's because like most other countries they've got a shortage of health care workers. They're not likely to tell us software engineers that we can com on in.
Canada is actively recruiting healthcare workers and it's apparently become quite easy to get people to move up. If I were a healthcare worker I wouldn't have to think about it for very long before having the U-Haul loaded up and ready to go.
https://healthcareinfusion.org is actively promoted on social media channels to assist with this, a project of the Canadian commentator, writer, and former national host and producer at CBC Radio Tod Maffin. BC allocated $5M to a marketing budget to do the same.
“Canada skilled workers program” are the relevant keywords for searches.
If you can’t find one that fits the work that you do, another option are visas that are non lucrative non working that are based on your investments, their income, including income from rental properties. Own a place in the US? Find a property manager, rent it out, visa secured (assuming monthly/annual income requirements are met).
The shortage of health care workers has a larger impact on the day to day of Canadians than a shortage of tech workers. However we would even define how many tech workers one "needs".
i dont think theres anything active, since canada has a bit of a glut of software engineers, but the big companies frequently put people who couldnt get an h1b in canada, so there must be some options
Run Qwen3-coder-next locally. That's what I'm doing (using LMstudio). It's actually a surprisingly capable model. I've had it working on some LLVM-IR manipulation and microcode generation for a kind of VLIW custom processor. I've been pleasantly surprised that it can handle this (LLVM is not easy) - there are also verilog code that define the processor's behavior that it reads to determine the microcode format and expected processor behavior. When I do hit something that it seems to struggle with I can go over to antigravity and get some free Gemini 3 flash usage.
We definitely saw some kind of non-linear step function jump in quality around the beginning of the year - it's hard to express how good Claude opus/sonnet 4.6 is now. However, I wonder if we're going to see the same kind of improvement from here? It's kind of like we got to the 80% point but the next 20% is going to be a lot harder/take longer than that first 80% (pareto principle). Also, as more and more code out there is AI generated it's going to be like the snake eating it's own tail. Training models on AI generated code doesn't seem like it will lead to improvements.
reply