Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
tornikeo
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
tornikeo
4d ago
"Everybody should slow down AI research except for me"
2.
▲
The Handshake Deal Protocol
(ycombinator.com)
1 points
by
tornikeo
15d ago
|
0 comments
3.
▲
by
tornikeo
17d ago
I use chatgpt work's VM as a CI/CD runner, in addition to my primary one on hetzner. It's a beefy 20GB 9CPU machine, why waste all that precious RAM? :)
4.
▲
A Modest Proposal for Anthropic
(tornikeo.com)
3 points
by
tornikeo
17d ago
|
1 comments
5.
▲
by
tornikeo
24d ago
lol. I was quite disappointed too
6.
▲
by
tornikeo
24d ago
This is absolutely nuts!
7.
▲
by
tornikeo
29d ago
> Deep integrations with Cursor No thanks. I want GitHub alternative, not Cursor with source control attached.
8.
▲
Localizing Slack
(slack.engineering)
1 points
by
tornikeo
1mo ago
|
0 comments
9.
▲
by
tornikeo
1mo ago
> I genuinely think Dario is a well intentioned, intelligent dude, but I think him and Anthropic have a huge PR problem Is this what kids call rage bait these days?
10.
▲
by
tornikeo
1mo ago
> Hi! > Hello! What would you like to work on today? > I want to talk about quantum gravity > Alright, let's get started. What would you like to work on today? > List and explain the current mainstream quantum gravity th
11.
▲
by
tornikeo
1mo ago
Claude Code has become the largest noobtrap I've ever seen. Move, try something else for a change. Codex, Pi, OpenCode, DeepSeek's harness all great harnesses with zero bullshit or drama.
12.
▲
by
tornikeo
1mo ago
Yet another (accidental) win for the open-weight models.
13.
▲
Fairy Ring
(en.wikipedia.org)
17 points
by
tornikeo
1mo ago
|
0 comments
14.
▲
by
tornikeo
1mo ago
https://peachjam.dev/ A low-cost vibecoding website for non-developers. Still in alpha.
15.
▲
by
tornikeo
1mo ago
Humans are just so damn different in their tolerance to wasting time. I can't take feeling like I wasted 10 seconds of my time on an ad even when watching an video for recreation.
16.
▲
by
tornikeo
2mo ago
SoTA AI will move well past 1TB+ memory requirements in 1 year or so. Also, you will be able to play with much more competent models locally. They will still feel like children compared to the adults living in the SoTA region.
17.
▲
by
tornikeo
2mo ago
...if done correctly. You can reverse engineer websites at a breathtaking speed if you do it correctly. PeachJam.dev took me approximately 2 months to make, for instance. If I had access to Sol from the start it would likely just take 1 mon
18.
▲
by
tornikeo
2mo ago
https://peachjam.dev/ It's like lovable but much more affordable. Built on Cloudflare + Hetzner. Still in beta.
19.
▲
by
tornikeo
2mo ago
The invisible hand of the free market is working wonders.
20.
▲
by
tornikeo
3mo ago
Surely the electromagnetic radiation from iPhone must be disorienting the storks.
21.
▲
by
tornikeo
3mo ago
All they had to do is to keep their mouths shut.
22.
▲
by
tornikeo
3mo ago
I really need this as an API. Turns out, to use Claude Agents SDK, you need to have a vision enabled API. If Deepseek API could see, it can fully drive Claude Code and Claude Agents SDK. A project I'm working on relies on a Claude-in-C
23.
▲
by
tornikeo
3mo ago
Nobody's stopping you from taking advantage of that shortage :) Just don't forget to turn your garage EUV machine on and off if it starts glitching.
24.
▲
by
tornikeo
3mo ago
For now codex. Didn't manage to get others to work well. And fully aware that I'll have to move to another thing after OpenAI enshittifies this as well.
25.
▲
by
tornikeo
3mo ago
That's a baseless claim that borderline reads like shilling. Do you have any proof of that you wrote there?
26.
▲
by
tornikeo
3mo ago
I moved off Claude Code 3 months ago. That decision keeps getting better and better as time goes on.
27.
▲
by
tornikeo
3mo ago
YMMW but on my local tasks Opus 4.8 is nowhere near close to gpt 5.5. For the lack of a better word Opus is just soo damn lazy.
28.
▲
by
tornikeo
3mo ago
This was an insightful comment, thank you for writing. And here's to hoping that the exponentially growing technological capabilities will allow curing death in our (short) lifespans. By God we need that to counteract the ageing popula
29.
▲
by
tornikeo
3mo ago
Tobacco companies should've just announced that "This new tobacco we made is super good, but too dangerous to release, so we are smoking it ourselves and giving it to only the select C-level smokers"
30.
▲
by
tornikeo
4mo ago
That's as easy as making new Vite. :) Which is hard, not easy but my point stands.
More ›