Or perhaps this is a new reality where for each current problem we will see myriad solutions because of how easy it is to put together a prototype with agents. 3 years ago having a sandbox was also an important problem to address, but we didn't see so many attempts.
It's a different level of isolation, worktrees help agent work on different code repository in parallel but things get wonky once you consider processes and environments variables
The way I do it is each worktree gets its own .env, with dev credentials to whatever it needs access to, and my coworkers have theirs set up to have an isolated Postgres database per worktree, but I didn’t bother with that in my setup— maybe I will, though. It’s a simple script to create a worktree and properly prepare it.
You're tragically misinformed; it isn't. Several metrics are actually growing exponentially. But if you want emprical information, you can just have al look at the nature of the late AI incidents.
Ironically, many benchmarks being maxxed out, and quite quickly, so new ones have to be created.
You prove your own point no? You are asking for a benchmark to prove AGAINST ASI. Surely the burden of proof for such a scientific fiction concept should be the other way around.
It was not solved. ~OpenAI~ Buckmaster and Alpöge found one (or a few) singularities in the forced version of the Navier-Stokes equations. Then magically 2 weeks later OpenAI found them too. Again, I am not saying this is not a great feat. I am just saying that everyone should be a bit more careful when making statements about RSI.
As the sibling comment points out, that is incorrect: they solved a smaller, simpler problem, and OAI solved the actual Millennium Problem. But even if they had solved the actual problem and OAI stole it by digging through chats, that hardly supports the "AI has stalled" thesis; Claude played the major role in creating their blowup to a different problem.
How, exactly, does "it was Claude that solved Navier Stokes, not ChatGPT!" get you to "AI has hit a wall and stalled"? That's, not to put too fine a point on it, incoherent, and is just noise thrown into the discussion to avoid grappling with the fact that AI continues to rapidly improve.
Buckmaster and Alpöge has found a forced finite-time singularity for the 3D incompressible Euler equations (and two other types) building on the work by Diego Córdoba and Luis Martínez-Zoroa with the assistance of Anthropic and OpenAI models. Then OpenAI found a forced finite-time singularity for the Navier-Stokes equations.
TL;DR Buckmaster and Alpöge haven't solved Navier-Stokes blow up.
How information can get so distorted when it's trivial to fact check?
I don't agree with "readable," though. For a simple set of key-values, yes. Once you get into complicated structures, other file formats express the semantics much better.
I think there are many facets to readability. For example, YAML is frequently held up as much easier for humans to read than JSON, but even after many years of reading and writing YAML in various domains, no other format causes me anywhere near as much trouble.
With JSON, I can format the string as I feel makes sense for that specific data structure, while YAML forces me to into specific indentation patterns, and it still causes me to question every time whether the dashes in arrays should be indented or not. And since people frequently template YAML, the strict indentation has caused a bunch of issues for no good reason, including production outages (sure, shouldn't happen with good practices, but there are so many places without good practices!).
Because JSON syntax generally has only one way to represent each data type and almost every data type uses explicit start and end signifiers (except floats and bools, both of which are short and clearly stand out), it's easy to know what the context of every character is both while reading linearly, and when jumping to specific points. Meanwhile YAML has multiple ways to express almost anything, and frequently the only way to tell the current context is to read ahead before jumping back. This is especially terrible for strings due to optional delineation, because almost every bare text could be a keyword (as demonstrated by the Norway problem).
I know this is not most people's experience, but that's because readability is subjective.
If you have a complex configuration you are doing it wrong... Spend some time thinking about you really want and turn it into a simple key-value setup. JSON should feel like it is overpowered and bloated for your needs. (I'd still use JSON because you can find a parser and editor that can handle it, but it should feel overpowered for your needs)
the connection I'm making is that both sit on the same interface: the CUDA driver api (module load, launch, memcpy).
the paper is about compiling kernels using safe rust instead of CUDA C++, what I did is reimplement the other side of it using rust for the calls from inside a VM get forwarded over vsock to the host driver via api-remoting.
reply