qwerpy

I very much want this to be true, I would shift as much of my workload as possible to local models. But is qwen 3.8 really good enough to do reverse engineering work? Is it because you have many sessions running at once, so it can "brute force" its way through problems?

They're the safest cars on the road and they drive themselves. If you want to maximize your family's safety, you put them in a Tesla. For me, the stress of driving is completely gone now. Traffic jams, slow drivers, complicated turns in unfamiliar areas, none of that bothers me anymore.

If Chinese cars were available here for cheaper, drove themselves just as well, and were similarly safe, I'd consider them.

Claude's insistence on protecting me from myself has already caused me enough frustration that I'm lowering my subscription tier and exploring if other models will work for me instead. The agents work in a VM I've set aside for them or in the cloud, the code & assets are backed up and in source control, and I don't care if the VM gets entirely destroyed because I can rebuild it. I've written this down for the agents. And yet the classifier keeps blocking agents from doing innocuous things like cloning an additional repo on-demand.

Having a "EVEN_MORE_DANGEROUSLY_SKIP_PERMISSIONS" flag is exactly what I'd use. Maybe Anthropic saw the constant profanity in the prompts I've been sending after getting refused.

Yeah, agreed. I do the same. I only contribute smaller patches that definitely fix bugs, and not big vibe coded sprawls of code that happen to work for me but who knows what it breaks. For example I started using opengym which is definitely someone’s vibe code project and then grafted SSO on top for my own needs, but there’s no way I’d submit that upstream. That’s fine. I got what I wanted out of it. Things will adjust but for now just enjoy the ride :)

Seems like a good use for AI. If people can use it to port video games from one system to another or jailbreak consoles, surely "get this library/program compiling on this device, then get all the tests passing, here's root ssh and an internet connection" could work pretty well in most cases. I wouldn't have bothered wasting my time in the past, but now, sure why not? It's not my time anymore and AI can turn the crank until it works.

Haha. I appreciate the intent behind this. HN will hate and downvote it of course, but we need constant reminders that we are emotional, take sides just like everyone else, and selectively get outraged about things.

The downvotes on your comments are weird. Maybe coming from an emotional place. I get it, I wouldn't want to be constantly reminded that the career I sacrificed so much for is a dead end. But it's true, at big companies there may be a ton of items in the backlog but none of them bring in more money. Having a bunch of idle engineers sitting around wastes money, and LLMs allow more engineers to do less. Why would a responsible VP keep around expensive employees who don't contribute to revenue?

Not a truck guy by any means but I wouldn’t want any other truck’s bed after having the cybertruck’s. The motorized tonneau cover and sloped roof adds a lot of volume to what a normal truck bed provides. It’s made the difference from having exposed camping/lake gear to having all of it protected and sheltered from the elements. Now I see my friends’ truck beds and rib them a bit about how petite their storage space is.

"Best available public data source" isn't a good reason to insist on using irrelevant data to draw conclusions like "top 5 models by token consumption are Chinese/open models".

Strange, I lived in very blue states (WA and CA) and they definitely punish EV owners with large annual surcharges on registrations. To offset “not paying gas tax”. Money is money, red or blue, they will all take as much as they can from you.

Not sure why online forums attract this kind of permanently grumpy person whose only contributions are complaints and nitpicks and snark, but I’m happy to shame, downvote, and flag them every time.

Claude is happily reverse engineering for me, a recently sunsetted mobile game and an old PlayStation game. If it doesn’t moralize at you and refuse to do it, it’s easier to use the smarter model on the subsidized subscription. I’m doing as much as I can now before they start refusing.

It'll be hard, when I'm never causing any accidents or speeding or running reds/stop signs. Maybe they could get me for inattentive driving but that's what the tinted windows are for.

Your information seems to be outdated, unless every single street I’ve been on for the past year is “vetted”. All the other (consumer) car brand’s self-driving systems do have that limitation though, maybe that's where the confusion came from. Tesla is self-driving 98% of my mileage, with the 2% being when I take over to choose a specific parking spot or I’m towing/carrying something big.

Don’t know much about how it does in the rain since moving to SoCal but in clear weather it hasn’t come close to hitting anyone yet. Even proactively warned me a few times while I was driving. Maybe I should let it take over in parking lots too :)

This is good advice and I wish I had read it before today. I use the state of the art model (Opus 5.5) and telling it with words how to interact with my issue tracker and my repo still doesn’t work. Mostly because of its overly restrictive auto-mode guardrails and brittle command whitelisting, but also a lot of “oops, I didn’t read the doc that you clearly tell me to read at the top of agents.md, I’ll make an ineffective note so this never happens again (it keeps happening)” The solution I’m using now is various scripts that encapsulate almost everything the agents usually need to do and just whitelist those.