Hacker News: qwerpy

New comment by qwerpy in "Tesla 'Cannot Guarantee' Roadster Reservation Holders Will Ever Get a Car"

qwerpy — Sun, 18 Oct 2026 17:09:50 +0000

At this point if you’re able to afford FSD and you choose not to for politics, you’re just making your life worse and less safe for silly reasons. Fine by me as I get driven around in my tank, except for the fact I have to be inconvenienced by human-caused wrecks and my insurance is higher to subsidize them.

New comment by qwerpy in "Zuckerberg's Meta Takes Down No Kings Instagram During Protest Day"

qwerpy — Sun, 18 Oct 2026 02:37:10 +0000

The bright minds here are busy posting Trump inauguration photos, trying to “gotcha” free speech advocates, and talking about prison time for tech CEOs. Quite the imagination these guys have!

New comment by qwerpy in "DeepSeek-v4.1-Flash-1017"

qwerpy — Sat, 17 Oct 2026 21:14:34 +0000

Yeah, I have the same question. I'd like to incorporate cheap & fast models like DS into my workflow to reduce my reliance on OpenAI/Anthropic, but after trying several such models and having to have a frontier model review/fix/redo everything multiple times per task, I gave up and simply have the frontier models (Sol/Opus) do everything. All of this fits pretty comfortably in my monthly subscription.

Maybe the equation changes if I exceed my subscription usage, and need to fill the gap with something else. But it seems like a huge waste of your time to go out of your way to use a cheap model if you have the tokens to spare in your subscription.

New comment by qwerpy in "Tesla Roadster"

qwerpy — Fri, 16 Oct 2026 19:35:15 +0000

Financial flop, maybe. I put down the pre-order fee on day one, then paid foundation series price for delivery and 2 years later it's still objectively the best car I could have for my family (space for toys, self-driving, safest truck on the road; not every family has these requirements). If they stop making it for financial reasons, that's fine, I'll have gotten my years of use out of it and hopefully someone else will have figured out how to put FSD on a production car.

New comment by qwerpy in "Using AI to reverse engineer hardware"

qwerpy — Fri, 16 Oct 2026 15:31:09 +0000

Yeah I’m rushing to reverse engineer some of my favorite games now when I have tokens to burn. Document the addresses, functions, break the encryption, unpack all assets. There’s no way opus keeps this functionality for long. Then I can have a dumber model continue the work or a smarter model with guardrails reassemble the pre-extracted logic and assets.

New comment by qwerpy in "What I self-host"

qwerpy — Thu, 15 Oct 2026 22:31:17 +0000

That’s what I’ve been doing. Managing this stuff by hand is awful and I’m not a sysadmin who is good at it. I just added single sign on to my home network for a wide variety of various services. Some of them didn’t even support SSO but a quick AI patch fixed that. There is no way I would even attempt that by hand let alone succeed.

I saw statements to the effect of “this would be nice but it’s a lot of work” several times in the post. I don’t understand why that has to be a blocker anymore. All this tedium is gone.

New comment by qwerpy in "Fire at Starlink site in Bulgaria after suspected sabotage in Poland"

qwerpy — Thu, 15 Oct 2026 01:38:19 +0000

“It’ll never work” -> “someone else can do it but not him” -> “if he does do it, it’ll bankrupt him” -> “well he did it late” -> “it wasn’t actually him” (we are here now) -> “fine but i still don’t like him”

The [flagged][dead] who replied to your earlier comment is an example of someone who reached the final, terminal stage.

New comment by qwerpy in "AI coding agents generate more code, but not more software"

qwerpy — Mon, 12 Oct 2026 23:18:00 +0000

My homelab has many fancy new things like a self-hosted gym app and many more metrics and dashboards in my home assistant instance, both of which I use on a daily basis. It also now runs a linux VM that is an always-on agent workspace, and various emulators & decompilers and their MCPs. I, like apparently everyone else, am also reverse engineering games and mashing them up for my own personal enjoyment. I've modded a few games for fun. I've done a security and utilization audit of my server and saved hundreds of dollars in server upgrades I would have bought if I didn't have AI identifying waste and optimizing things. I could have done all of those things before but I didn't, because of how annoying it is to have to dive so deep into the weeds to do small things.

People who have the technical skills and free time for these hobbies, and aren't having the times of their lives right now, are missing out.

New comment by qwerpy in "AI coding agents generate more code, but not more software"

qwerpy — Mon, 12 Oct 2026 23:09:47 +0000

Is he not allowed to use Claude to do the reviews? I've heard from friends that the engineers on the team are basically now just routers for Claude/ChatGPT.

New comment by qwerpy in "QGIS enacts moratorium on AI contributions"

qwerpy — Mon, 12 Oct 2026 15:01:44 +0000

Yeah I sympathize with this, but I’ve experienced the other side of this: I ran into a bug (an integration for home assistant has to be manually kicked every time HA restarts due to a bug in the integration) that my AI fixed with a one-liner. It works great for me. I submitted a PR which is languishing for weeks. No problem. My AI will just maintain a personal fork with my patch until my PR gets merged (or not). Previously it was way too much work maintaining your own fork, merging in upstream changes, and keeping your patch working.

I think this is going to be a common outcome from the fact that many people have AI agents that can one-shot fix and deploy. Maybe it’s not a bad thing for those people and the maintainers who each get what they want.

New comment by qwerpy in "QGIS enacts moratorium on AI contributions"

qwerpy — Mon, 12 Oct 2026 14:53:19 +0000

PMs are doing demos to leadership where their change “just works”. The devs and maybe even the dev leads may protest that due diligence wasn’t done, but leadership is looking at the engineering cost center, axe in hand.

New comment by qwerpy in "QGIS enacts moratorium on AI assisted contributions"

qwerpy — Mon, 12 Oct 2026 14:41:50 +0000

Better than remaining in denial about it like most people on here.

New comment by qwerpy in "The gap in shared understanding of LLM capability is widening"

qwerpy — Sun, 11 Oct 2026 23:24:30 +0000

Every time I use Google search AI for anything more complicated than a search query I feel that way. A paid model from pretty much anyone is much better.

New comment by qwerpy in "Bitwarden Dual License Model"

qwerpy — Sat, 10 Oct 2026 17:23:29 +0000

At the time I decided to go with regular keepass, it was because setting up OneDrive sync (directly to OneDrive, not depending on a mapped folder because I use as little MS software as possible) and browser extension was easier. And the cross platform UI on KeePassXC just didn’t look and feel good on windows.

New comment by qwerpy in "YouTuber Says Cops Visited Him After He Built a Flock-Style Camera to Track Cops"

qwerpy — Fri, 09 Oct 2026 22:46:29 +0000

> Am I missing something?

Yes you missed something big. A little too excited to get in the America dig, didn’t even bother reading the article to realize it happened in Canada.

New comment by qwerpy in "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"

qwerpy — Thu, 08 Oct 2026 20:59:08 +0000

Yeah. $100 for Claude just about gives me all the usage I want, as a more or less full-time hobbyist having it work in the background most of the day. I was trying to economize by having a local LLM, then Deepseek, then Cursor/Grok, and then I got a taste of Opus 5.5 and I simply cannot go back to having to carefully spec things out and double-check work. I just let it decide, Opus or Sonnet for the next task, and I get almost perfect results. Probably similar with OpenAI's models.

The token-equivalent monthly spend is > $5K+. If Deepseek's token cost is 20x cheaper, that's $250/mo, and I'd be spending a lot more of my brainpower babysitting it and getting worse results.

For business/team accounts that pay per-token, maybe I can see the "freaking out" being warranted on the part of the fronter labs. But as long as they're willing to subsidize their end-user subscriptions, I'm not going to move off of them until the alternatives are truly at their level.

New comment by qwerpy in "PS5 Jailbreaks Are Escalating at an Unprecedented Pace"

qwerpy — Wed, 07 Oct 2026 15:12:49 +0000

I had my LLMs RE a sunsetted mobile game and within a few days ($10 equivalent of my monthly plan) replicate the fight screen with nearly perfect pixel by pixel animations and visual effects. Heavy use of ghidra and a MCP. Then it installed an android emulator to build an automated compare pipeline between the game and my remake. When I don’t feel like babysitting it, I tell it to spend some tokens on the long backlog of RE items and it happily churns away at it. It’s so easy and cheap, anyone can do this by themselves, for themselves. No need to wait for someone else to do it.

A swarm of humans with frontier models could probably rebuild many mobile games in a month. Only limiting factor is the lawyers. I’m somewhat surprised the models don’t have guardrails for this. They probably will soon enough, but by then the Chinese models will be good enough to do it. The dream is to have local models that can do it.

New comment by qwerpy in "What Meta got right with Muse"

qwerpy — Sun, 04 Oct 2026 18:30:44 +0000

Disregarding the politics because that’s always subjective, it’s central to Meta’s business model to harvest as much of your data to show you ads. And they have decades of track record of doing that. Same with Google. Their products are thoroughly corrupted by ads because that’s where they make their money.

xAI, OpenAI, Anthropic, may dabble in ads but they are nowhere near Meta and Google.

New comment by qwerpy in "Tesla takes on $30B in credit as it approaches unprofitability"

qwerpy — Wed, 30 Sep 2026 06:09:36 +0000

Yeah I’d love a cybervan or model XL. Pretty much every family I know would get one. In lieu of that we all got model Ys / cybertrucks instead, and then we grumble about it when the in-laws come to visit and we need to drive two cars to take the family out. YL is fine but they needed to go further. Need 3 proper rows and storage space!

New comment by qwerpy in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

qwerpy — Wed, 30 Sep 2026 01:21:14 +0000

4x RTX 6000 running at full load with relatively cheap electricity ($0.25/kWh) is around $450/month.

I did the math and decided it’d better to pay for tokens than to buy the hardware and generate them myself.