grim_io

19 ms·

451.
by grim_io 1y ago
One man's AI hype is another man's tangible productivity boost and/or UX improvement.
452.
by grim_io 1y ago
Relevant: "Why are anime catgirls blocking my access to the Linux kernel?" [1] [1] https://news.ycombinator.com/item?id=44962529
453.
by grim_io 1y ago
I'm considering removing serena MCP, since cc got better with its own tools. The new /context cc command is great for visualizing what uses how much of the context. On the other hand, I'm curious about dagger's container
454.
by grim_io 1y ago
I've tried some of those "frameworks" for claude code, but it's difficult to measure any objective improvement. I tend to lean towards them being snake oil. A lot of process and ritual around using them, but for what? I
455.
by grim_io 1y ago
If you move into the datacenter, you will benefit from the volume discounts on energy ;)
456.
by grim_io 1y ago
Such genius. Next, Musk will revolutionize manufacturing by using humanoid robots. Of course, they will be individually operated by a person.
457.
by grim_io 1y ago
I've got the solution! Digitize your brain and move into the data center. Can't recommend the show "Pantheon" enough.
458.
by grim_io 1y ago
Unless they get Thanos snapped out of existence, I would probably just switch to inferior local models. Going back to pre-LLM is not an option for me. Not because I can't, but because I don't want to. How are you people using AI?
459.
by grim_io 1y ago
I have never used LLM's to write essays, so I can't comment on that. What I can comment on is how valuable and energizing it is for me to cooperatively code with LLM's using agents. I find it sad to hear when someone finds th
460.
by grim_io 1y ago
Are you against taxes in general, or what's up? I have no problem with it. I would even argue that the EU is under funded.
461.
by grim_io 1y ago
What means? Are you unaware of the massive Russian government sponsored misinformation and political influencing? And no, whatever the US or the EU do, it is not the same.
462.
by grim_io 1y ago
I would say there is a much stronger incentive for anti EU propaganda nowadays. Certain state level actors would love some more anti EU sentiments right now. On the other side, who the hell would pay for pro EU propaganda?
463.
by grim_io 1y ago
I've tried to switch to Dvorak, but after a year I was just tired of the universally shitty default keybinding experience. Either customize every app to better match the layout, or live with the afterthought UX. Oh, and gaming. Better
464.
by grim_io 1y ago
I'll take my good old backroom deals EU over the shitcoin peddling US officials any day of the week.
465.
by grim_io 1y ago
Nah, don't agree that struggle adds much. I'm in the 10<x<20 range of experience. People will learn if they find the lesson valuable and interesting, no matter how much struggle and time wasting was involved.
466.
by grim_io 1y ago
This self-sabotage of self-sabotage is not something I can do. My brain has too much agency for its own good. It would not let itself be constrained in its pursuit of scrolling bliss.
467.
by grim_io 1y ago
Can I trouble you to give me a hint where this could be configured? :)
468.
by grim_io 1y ago
Opt-in would be much better, of course. To put it in perspective: google won't even give you an option to opt out. If you pay for Gemini as a private user and not as a corporation, you are fair game for google. Now, neither option is g
469.
by grim_io 1y ago
Sorry, I meant a very fancy next token predictor :)
470.
by grim_io 1y ago
I think you are selling yourself short if you believe you can be replaced by a next token predictor :)
471.
by grim_io 1y ago
Sorry for not making my point clear. I'm arguing that the category of the problem matters a lot. Chess is, compared to self-driving cars and (in my opinion) programming, very limited in its rules, the fixed board size and the lack of &
472.
by grim_io 1y ago
There is a reason chess was "solved" so fast. The game maps very nicely onto computers in general. A grandmaster chess playing ai is not better at driving a car than my calculator from the 90s.
473.
by grim_io 1y ago
Sure, you can upscale a badly compressed jpeg using ai into something better looking. Often the results will be great. Sometimes the hallucinated details will not match the expectations. I think this applies fundamentally to all of the LLM
474.
by grim_io 1y ago
The jpeg compression argument is still valid. It's lossy compression at the core.
475.
by grim_io 1y ago
No matter how much better they get, I don't see any actual sign of intelligence , do you? There is a lot of handwaving around the definition of intelligence in this context, of course. My definition would be actual on the job learning
476.
by grim_io 1y ago
Maybe it's just the nature of being early adopters. Other fields will get their turn once a baseline of best practices is established that the consultants can sell training for. In the meantime, memes aside, I'm not too worried ab
477.
by grim_io 1y ago
One man's session hijacking is another man's unofficial third party client support.
478.
by grim_io 1y ago
We don't even need to compare it to the most expensive subscriptions. The $20 subscription is far more capable than anything i could build locally for under $10k.
479.
by grim_io 1y ago
That's quite a bold statement to assume that the reader is probably using uv wrong. I'll give some unsolicited advice while on topic: On Linux, on your development machine, don't use your system python for your venvs. Your
480.
by grim_io 1y ago
Claude Code is currently the best agentic coding tool available. From my own experience, I can say that the only thing that can improve upon Claude Code is something that can provide better tools to Claude Code. I'm very happy with ht

More ›