HNHacker News
TopNewBestAskShowJobs

taytus

5,075 karma · joined May 23, 2011

submissionscomments
taytus··on Claude Opus 4.8
Incremental gains compounds.
taytus··on Green card seekers must leave U.S. to apply, Trump administration says
Signaling.
taytus··on I'm going back to writing code by hand
Nice Fiction story.
taytus··on DeepClaude – Claude Code agent loop with DeepSeek V4 Pro
Kimi feels almost limitless. I have the $40/month plan and I've never ever remotely close to hitting the limit. Using opus as the orchestrator.
taytus··on Tell HN: Claude 4.7 is ignoring stop hooks
Skills are not like hooks. Skills can and will inevitably be ignored.
taytus··on Tell HN: Claude 4.7 is ignoring stop hooks
Hooks are hard stops. In theory the model must respect them, unlike Claude.md or agents.md so yeah, it helps a lot.
taytus··on Tell HN: Claude 4.7 is ignoring stop hooks
Yes an no. Some skills are very very tuned to our own workflows. The model providers may come up with some similar alternatives but not always. Also, sometimes you need a solution now and not in three months.
taytus··on I cancelled Claude: Token issues, declining quality, and poor support
I don’t have the prompt at hand but basically I told Kimi (paraphrasing): I have these Claude code skills, and I know it uses different tool calls than you but read them and re-write them as your own tools.

I also created a mini framework so it can test that the skills are actually working after implementation.

Everything runs perfectly.

taytus··on I Cancelled Claude: Token Issues, Declining Quality, and Poor Support
Straight from them, but I know other providers like io.net can be faster but I like to directly support the project.
taytus··on I cancelled Claude: Token issues, declining quality, and poor support
I'd recommend Kimi k2.6 for your use. It is an excellent model at a fraction of the cost, and you can use Claude Code with it.

I did a 1:1 map of all my Claude Code skills, and it feels like I never left Opus.

Super happy with the results.

taytus··on DeepSeek v4
>I asked DS itself and it denied this

Bro, seriously?

taytus··on An update on recent Claude Code quality reports
“after evals and dogfooding” couldn’t have done this before releasing the model? We are paying $200/month to beta test the software for you.
taytus··on An update on recent Claude Code quality reports
They should do a similar report about their communication team. This was horrible mismanaged.
taytus··on Claude Design
Anthropic is making the exact same mistake OpenAI did: lack of focus. Release, release, release, all good apps, all full of bugs. This is a massive mess and people aren't talking enough about how bad all of this is.
taytus··on Claude Code Routines
You are arguing something different. My point is that they must apply these restrictions. Do I think they could have calculated their growth a little better? Yes, of course, but hindsight is 20/20.
taytus··on Claude Code Routines
Like or not, openai isn't having the same compute strain, meaning this was predictable.
taytus··on Claude Code Routines
They are not dumb restrictions. They just don't have the compute. That is the dumb part. Dario did not secure the compute they need so now they are obviously struggling.
taytus··on System Card: Claude Mythos Preview [pdf]
That has also been my experience. And if Mythos is even worse, unless you have a significantly awesome harness, sounds like pretty unusable if you don't want to risk those problems.
taytus··on Folk are getting dangerously attached to AI that always tells them they're right
The stupidest people you know are getting the “you are absolutely right!!” Validation they do not need
taytus··on Iran-linked hackers claim breach of FBI director's personal email
Who said they are the enemy?
taytus··on Goodbye to Sora
How much money did they burn on this? And for what? Nothing?
taytus··on Cursor Composer 2 is just Kimi K2.5 with RL
YC is back at it again.
taytus··on Yann LeCun raises $1B to build AI that understands the physical world
He raises $1B, couldn't OAI, Google or Anthropic try similar approaches? Lack of funding isn't a problem those companies have. Why wouldn't they also spend $1B or 5 times that and outcompete (in theory)?
taytus··on Anthropic announces proof of distillation at scale by MiniMax, DeepSeek,Moonshot
This reads like AI slop.
taytus··on Gemini 3.1 Pro
Another preview model? Why google keep doing this?
taytus··on Claude Sonnet 4.6
Honest question: why would anyone use Opus instead of this? I’m doing web development, the whole shebang, and I don’t think I need Opus right now. I know it’s supposed to be smarter, but a 2%–5% improvement doesn’t seem meaningful, especially when it costs more than double and has only a portion of the context window.

Am I getting this wrong? I would seriously appreciate any clarification here.

taytus··on A sane but bull case on Clawdbot / OpenClaw
>You do understand that is who you’re competing with now right?

No. I'm competing with no one.

taytus··on A few random notes from Claude coding quite a bit last few weeks
Because you are not coding, you are building. I've been coding since I was 7 years old, now I'm building.
taytus··on Roam 50GB is now Roam 100GB
Fascism isn't just American. If you're going to be pedantic, at least be accurate. My point: supporting an American fascist normalizes fascism everywhere
taytus··on Roam 50GB is now Roam 100GB
It’s not “American politics” when a guy does a Nazi salute on live TV. So thanks for showing your political inclination.
← PreviousPage 2 of 20Next →