Creatine doesn't need astroturfing.. It's not <frontier AI lab>'s newest model turbo boost max pro extreme rocket 5.9. It's just a supplement that some people take.
I've heard that. I hope that the people who are IDE-free are just better AI-wranglers than I am because my experience has been that if I can get an agent to one shot something, it's fine but if I can't, the agents tend to make an absolute mess of spaghetti that doesn't actually do what it was asked to do IME.
But what about navigating the code by the call stack? I didn't know that GitHub has a way to do that. Or maybe I'm probably coming across as being dumb enough to be talking about still trying to have a mental model of what calls what.
I've never used breakpoint debugging, was always a printf debugger. And now an agent can do that loop for me.
Prompt is usually something along the lines of:
>I would expect the behavior of this to be [X] - instead I'm observing [Y]
And the agent will form hypothesis, place printf statements, compile, and scrape logs on loop - each loop ruling out hypothesis or narrowing down what portion of the code is responsible for the unexpected behavior.
It has been able to pin-point the exact line(s) of code responsible every time I've reached for it so far.
For what it's worth, generally speaking I read all of the code and keep it in my brain - I have some uncommon assets in that regard like a high reading speed and great memory. `git grep` is the other tool I use often.
I rarely find that the call stack is the limiting factor, to me, and I suppose I do something similar to what you're talking about but just in my head - I know where a file is referenced via imports, what a function does, and what the flow of control is like.
> Do you not need to use the debugger sometimes? Or can cc debug by itself
A key feature of AI coding assistants and coding agents is troubleshooting. It turns out that LLMs excel at pattern matching, specially when coupled with feedback signals. It turns out that troubleshooting represents just that. A few years ago people searched the likes of stack overflow to fix problems, and it turns out LLMs can do the equivalent of that much faster.
But tests show you if a bug is happening, they don’t help you understand the underlying cause of the bug. In a decade, you haven’t hit a compiler codegen issue, a silicon erratum, a race condition, or anything that required actually spending effort understanding the causal path?
I don't know what to tell you besides that I've found more, better bugs using comprehensive testing than I ever found with a debugger.
The sole exception to that is that, back in the very early days, troubleshooting IE6 really required a debugger. But everything else, from memory leaks to thread hang issues to deadlocks, testing is better.
Maybe? We dont really know this right? People have been saying this for 5 years now and the models are still getting better. The companies running the frontier models have already scraped everything on the web, but the models are still getting better, even if it's only marginally better, with each release. Maybe eventually some company will actually achieve AGI/ASI, who knows..
this was true a year ago but not so much anymore. You still have to supervise the agents, but they can write maintainable code if you keep an eye on it.
You're missing the point (not sure if you're just being dense on purpose...). If you're bank would just return the money then its not a good analogy. If someone gains root access to your machine, presumably they can do damage that can't be undone. In other words, to continue the bank analogy, they would take all your money and you would have no way of getting it back. Presumably, you would not be ok with this. And even if, for some weird reason, you were ok with that, 99.9% of all other people would not be ok with it.
Respectfully, I don't think they're missing the point. Banking, as an institution, has its flaws, but deposit insurance isn't one of them. These vulnerabilities exist whether or not they follow specific disclosure rituals, and systems should be deployed with defense-in-depth so that one privilege-escalation flaw is a recoverable event. Inventing tortured counterfactual analogies doesn't change the basic thrust of the poster's point: the account is insured, so getting drained by an attacker is not a fatal problem. Of course people should still take steps to prevent that from happening, but that doesn't mean prevention is (or should be) the only cure.
My point specifically is that some damage isn't recoverable if there's a vulnerability that gives someone root access. This makes the bank analogy inadequate in the first place. Im not trying to argue about whether deposit insurance is good or bad. Saying they would get the money back assumes the damage done to ones machine would be recoverable, which may not be the case.
My understanding is that FDIC deposit insurance only protects against bank failure, not fraudulent activity. Getting your account drained by an attacker may or may not be covered by a patchwork of other laws at various levels, and you could very well end up shit out of luck.
So you essentially trust the output of the model from beginning to end? Curious to know what type of application you're building where you can safely do that.
Edit: to clarify, I know these models have gotten significantly better. The output is pretty incredible sometimes, but trusting it end to end like that just seems super risky still.
I don't think that's unpopular, it is pretty well written. But the "I believe" section is extraordinarily hard to believe given Altman's history.
> Working towards prosperity for everyone, empowering all people
> We have to get safety right
> AI has to be democratized; power cannot be too concentrated
None of these statements, IMO, reflect his actions over the past 5 years.
> we urgently need a society-wide response to be resilient to new threats. This includes things like new policy to help navigate through a difficult economic transition in order to get to a much better future
I agree with this, but there is a near 0% chance of that happening anytime soon in the US. I think he probably is aware of this.
Just my opinion, but it comes off as very insincere.
To be clear, what happened is still awful and there's absolutely no justification for it.
I am taking a course through a website called neetcode.io
It is a clone of leetcode, designed to help you build intuition in a programmatic way, to learn the top 75 - 150 coding questions, common in interviews.
Each lesson comes with detailed video explanation, with practice problems. The practice problems too come with video solutions.
If you go to the main site, you will see a link to the different courses they offer and also a roadmap. The roadmap organizes the algorithms in a hierarchy, from simple complex, to help you reduce your blindspots, as you build your intuition.
I'm nearly complete with the beginner course and will move to advanced soon. For me personally, it works quite well, because I need a human to explain things to me in detail in order to understand the complexity.
reply