As I work with Claude more and gain a feel for its capabilities, I tend to run into 2 far less often, as I'll decompose my messages more for the current model limitations. The threshold also changes each release.
I think a reading of Roman history shows the failure modes of a senate so often that I wonder if it was ever supposed to work more than a few hundred years.
I switched from kimi to GLM for claws. I found it tends to be a stronger debugger when it comes to fixing openclaw issues with bash, being more thorough in finding the right root cause. Kimi tended to confidently misdiagnose. It's also much faster. The only downside is that it seems to cost an order of magnitude more than kimi.
The closest I attained to flow state vibe coding involved building UIs with claude 4.6 fast mode low thinking, a local TTS model (nvidia's), and hot reloading browser.
Arguably maybe I could have done better on higher thinking and then done more in parallel but it is more tiring.
If SWE Bench is public then Anthropic is at a minimum probably also looking at their SWE bench scores when making changes, I'd trust more a tracker which runs a private benchmark not known to Anthropic.
Yes, sometimes I have found it hard to sleep if I'm close to building something I want to build.
I think there is kind of a meme going around about multitaskers doing very well with vibe coding, and I can see it. Although, as someone who has the opposite problem, it can be tiring if I try to do more than two things at once.
I've found this to be true so far, junior engineers with AI can be super productive but they can also cause a lot of damage (more outages than ever) and AI amplifies the sometimes poorly designed code they can generate.
I suspect a lot of it best practices will be enforcing best practices via agents.md/claude.md to create a more constrained environment.
My take: openclaw should not run on a mac (even though looking at the skills it ships with it clearly was made to)
It should run on its own VPS with full root access, given api keys which have spending limits, given no way for strangers to talk to it. I treat it as a digital assistant, a separate entity, who may at some point decide to sell me out as any human stranger might, and then share personal info under that premise.
Just uses claude. I haven't tried it much but it seems to be what you're describing.
Openclaw uses pi agent under the hood. Arguably most of the codebase could be replaced by systemd if you're running on a VPS for scheduling though, and then its a series of prompts on top of pi agent.
Boris has been very open about the 100% AI code writing rate and my own experience matches. If you have a typescript or common codebase, once you set your processes up correctly (you have tests / verification, you have a CLAUDE or AGENTS.md that you always add learnings to, you add skill files as you find repeatable tasks, you have automated code review), its not hard to achieve this.
Then the human touch points become coming up with what to build, reviewing the eng plans of the AI, and increasingly light code review of the underlying code, focusing on overall architectural decisions and only occasionally intervening to clean things up (again with AI)
> Singapore’s free speech restrictions, whatever you think of them, no longer seem so far outside the box. Trump is suing plenty of people. The UK is sending police to knock on people’s doors for social media posts, and so on. That too makes Singapore more of a “normal country"
That seems like it should make Singapore _more_ cool, at least my personal theory is that this changed a lot of perception of China (at least in some parts of gen z social media, "it's a very Chinese time").
Hmm, I once transited in Heathrow in a return flight from europe to the US and had to go through Heathrow security for whatever reason, where they subjected me to liquids rules way stricter than either my source or destination did.
E.g. 1 day use contact lenses and prescription creams all having to fit in a tiny plastic bag. So I'm happy for this change.
Afaik the apps don't have to ask you, they could just request the OS-level permissions. They don't do that because if you reject the request at the OS level, they can't request it again, you have to go to the Settings app to enable it and it's harder to do. So apps prefer to just nag you again and again until you say you're ready.
2M units globally, so maybe generously 1M in the US (assuming you're from there but multiply by some factor proportional to your country's consumeristic tendencies), divided by the population is only something like a .3% rate of ownership. So not quite as prevalent as gopro, which has sold something like 35M in the US over the past 10 years [1]