I don’t think mapping tooling to ability makes sense here, particularly when the “advanced” tools here just abstract more away, though I agree that Cursor is terrible. So many useless windows.
You can paste this exact response under your other post, as well. Emotionally compelling, scientifically weak. Worth nothing.
But also, what you wrote is just wrong. There are plenty of measurable, significant, ood effects in the last few decades on the climate and its impact.
Here’s one study onglacier retreat over the last 20 years.
Author clearly doesn’t know the field well at all. First few paraphrases reveal this. Opening sentence: I’ve been in the AI space since ChatGPT first dropped.
Everyone is allowed to have an opinion, but that doesn’t mean they’re all worth listening to. Unfortunately, right now, all of those opinions are about ai.
That’s just playing semantics. Nobody is talking about, “objective facts” or need define them here. If the step time is measured in days, and your model takes years to train, then it will never get trained to completion on consumer hardware (the entire point).
But I’m responding to the rescue mission comment, which, since Vietnam, have overwhelmingly employed helicopters (Huey’s then, Black Hawks today). But machinery aside, the larger point is that air operations will likely go worse here than they did in Vietnam, unfortunately for both sides.
…against the viet cong, where the biggest risk was the pilot getting pierced from small arms fire (in addition to the helo going down from pilot error). Quite different from the anti-air weapons modern day Iran possesses.
Fair point, I just think that example is better off running on a cron job than using compute from llm inference, (though that will become negligible over time, anyways).
I see this critique about autoresearch online often, but I think it’s misplaced.
Here’s a use case that may illuminate the difference, from my own work at Nvidia. Im currently training some large sparse autoencoders, and there are issues with dead latents. Several solutions exit to help here, such as auxk, which I can certainly include and tune the relevant params as you describe. However, I have several other ideas that are much different, each of which requires editing core code (full evaluation changes, initialization strategies, architecture changes, etc.), including changes to parallelism strategies in the multi-rank environment I’m using. Moreover, based on my ideas and other existing literature, Claude can try a number of new ideas, each potentially involving more code changes.
This automated run-and-discover process is far beyond what’s possible with hyperparam search.
I agree with the sentiment that there are use cases for web scraping where an agent is preferable to a cron job, but I think your particular example can certainly be achieved with a cron job and a basic parser script. Just have Claude write it.