I'm not sure this is the issue. I asked Claude Code a simple question yesterday. No sub agents. No web fetches. Relatively small context. Outside of peak hours. Burned 8% of my Max 5x 5hr usage limit. I've never seen anything like this before, even when the cache is cold.
You don't. Most of the time (after the first prompt following a compaction or context clear) the context prefix is cached, and you pay something like 10% of the cost for cached tokens. But your total cost is still roughly the area under a line with positive slope. So increases quadratically with context length.
Yeah, the viz for polar quantization is straight up nonsensical. Okay, so some colors are converted into clocks and then into a bigger box with a pink box inside of it. Got it. Even understanding what polar coordinates are doesn't help you make sense out of it.
Yeah, and some parts of the article are just bizarre:
> Instead of looking at a memory vector using standard coordinates (i.e., X, Y, Z) that indicate the distance along each axis, PolarQuant converts the vector into polar coordinates using a Cartesian coordinate system. This is comparable to replacing "Go 3 blocks East, 4 blocks North" with "Go 5 blocks total at a 37-degree angle”
Why bother explaining this? Were they targeting the high school and middle school student reader base??
It would, but only once you have fallen under the spell of thinking meaning and authority was to be found there. Consider your zone has been flooded maybe?
The Claude Code TUI app is pretty solid. I use it heavily and I get great results from it. But with the mobile app, Claude Code remote is basically unusable (weird disconnect bugs) and Claude Code cloud has issues as well (UI hides approval confirmations; must reconnect to see them). So yeah, I imagine what you're saying is true. There are at least some major gaps in their QA process. It's ironically a pretty convincing case to keep humans in the loop. It's honestly shocking to me that those features were actually shipped in their current state. You run into the problems immediately.
I have a very different experience. Claude code tui is the worst tui I have ever used. How is it possible that an inactive tui regularly eats 8gb of ram, has freezing issues and rendering issues?
If I wasn’t forced to use it I wouldn’t as there are better options available.
I agree with you about the Claude Code TUI. I switched to it weeks after it was released. The browser interface is great for quick chats and talking through ideas/concepts, but not for coding. What I love about the TUI is that it can see all your repos as once so it has the full picture all the time. You can't get that with the web version.
i.e. Combing through public forums on the internet looking for evidence of thought crime, however, is fair game. The Trump admin will undoubtedly use tools like this to compile a list political enemies or undesirables, which they will then use to harass people or selectively restrict individual rights. They're already doing this and this is just going to make it easier for them.
I don't think it will feel even remotely tolerable in the US. I've been heavily critical of Trump on a regular basis on the public internet ever since he showed up 10 years ago. I doubt a government surveillance AI would miss this. Of course, there are probably millions of people like me, but given the behavior of the government recently, I really have to wonder what they might do to people like me once we've been put on a list.
I don't understand this. I use agentic coding to do things more quickly. And it's not just toys. I end up with software that both works and is useful. Assuming AI models powerful enough to drive that process continue to be available, why would I stop doing it?
Th is isn't discussing individuals, it's discussing trends as a whole. There are still plenty of makers getting value out of 3d printers as well, but it's not everyone like we talk about everyone becoming a software developer with vibe coding.
Software is just a collection of functions. Some input returns some output.
That is what people do every day with LLM. When they ask LLM to do something, they are being software developers without them or you even realizing it. But what are they doing but building something with the LLM that takes digital input and returns digital output. It is software. Summarize this email for me gpt. That is a tool.
reply