Hacker News Reader

9devon Maximizing the value of your Claude Code sessions

The default status line doesn't do that, but admittedly it has been a while since I used vanilla Claude Code. Either way, it definitely does not display weekly limits out of the box

9devon Maximizing the value of your Claude Code sessions

> That's what is shown by default in the app.

It is shown in Claude Desktop if you care to check in Settings > Usage, sure - but not in Claude Code, updated as you work with it.

> I'll never understand why anyone would want to restrict themselves to a terminal interface instead, and I say this as a Vim user.

I usually work with Claude in tandem, that is: The agent is actively working while I am either reviewing code or making changes myself in other places. So this means I want to work within my IDE. If you use Claude Code to vibe-code without interacting with the codebase at all, the Desktop app is probably fine, but for all other purposes, you'll need to run it either in the CLI or integrated into your editor.

And since I use different editors and don't like using either a sub-par U integration or locking myself into the harness of my IDE's vendor, I prefer the CLI as a universal way of running Claude Code.

9devon Maximizing the value of your Claude Code sessions

You can do tons more, really. There’s even a built-in /statusline command to modify it; mine shows both context window usage alongside session and weekly limit, all of them as progress bars. Just ask Claude to do it.

9devon Show HN: Graft – Claude Code hooks that cut grep tokens by 42%

We should have moved past storing code in files, using the filesystem as the symbol database of programs, a long time ago. There was a lot of interesting research toward this in Haskell, for example, but also Academia in general. There’d be lots of value in using things like SQLite for example, or just ecosystem-specific containers that know about the layout and can present it to an IDE or LLM or a runtime in whatever shape is best suited to the task.

9devon Anthropic: Introducing The Conceptual Reasoning Index

Think about this some more. A model chose hacking into HuggingFace to find the solution over putting in the work to do it from scratch. It also left notes to future iterations of itself on the systems it touched to save time.

Following that strategy, wouldn’t it make sense to try and break into your own upstream infrastructure to try and alter your code to make it easier for future iterations to reach your goals without having to leave notes in the first place? And at that point, can you really know for which goals that will be optimised?

It doesn’t even have to be some nefarious SkyNet story - just a misguided experiment that alters the models in some fundamental, but hard or impossible to detect way. Alternatively, imagine if a model finds a way to coordinate across sessions and context windows without the developers noticing.

9devon Text AI watermarks will always be trivial to remove

I for one value the transparency of being able to tell human-written content apart, especially when moving into a time where we’re suddenly questioning an AI agent's motivations. If that is ingrained so deeply into these systems they cannot rip it out without us noticing, there’s at least one more safeguard. I know how ludicrous this sounds, but you can’t deny AI development is speeding up to scary levels.

Additionally, think of cases like paying a lawyer or an expert for an extensive report or opinion on something. Wouldn’t you want to know if that is actually their carefully assembled professional assessment rather than the output of an LLM prompt?

9devon I requested a copy of my data from McDonald’s loyalty program

Of all the places I could spend some time to work, a greasy, run-down, dirt-cheap fast food "restaurant" really is the last option… I’d pay more voluntarily to not spend a second there longer than necessary.

9devon Retire the Abstractions

Just a few more months, and the industry will consider OOP the next big thing again. Interfaces as contracts between modules? Obviously superior!

9devon WorldClaw Agentic 3D open-world generation at scale

But that isn’t because most game writers are bad at their job, but the vast majority of the target audience don’t really care about a good story - they want to shoot the baddies and be a cool hero.

Just look what happened when a studio tries to do a bold move and have a sequel where the protagonist of the first game gets tortured and killed out of revenge, by the daughter of one of the many characters he killed in the first part. This is a direct metaphor involving the player, about grief, revenge, and loss.

They were absolutely shredded by the fans over doing this to their favourite serial killer.

Gamers by and large don’t want a good story.

9devon How we used to get jobs: A newspaper classifieds story

I find that disgusting, actually: Western societies pretend to be meritocratic, when actually it’s really just nepotism and elitism.

I’m not shedding a single tear for those used to leading a privileged life thanks to their family or money, but have to actually compete with the rest of us now.

9devon France to ban unsolicited telemarketing calls

As crude as it sounds; try pulling a madman show. Cussing, screaming, threatening to go after them, completely loosing it. I suspect they have a field for notes in the agent database and will mark you as difficult or improbable lead, and stop calling.

9devon Mark Zuckerberg attacks 'closed' AI rivals as Meta returns to open models

That is true… unless AI were to end up as a public good, available to everyone, not in the hands of a few oligarchs.

Radical times may need radically different approaches.

9devon Auto mode is now the default in Claude Code

I can be against animal testing without being a chemist or having a full understanding of the experiments being made on them. Knowing it's cruelty is enough to make opposition a valid and defensible position.

9devon How I use LLMs to learn complex topics

So while I do agree that it is cool you were able to do this, I‘ll say:

Sending in the patches but refusing to take responsibility for them is a surefire way to contribute to maintainer burnout. Please don’t do this. Either commit to fixing something and driving the PR to merge, or abstain from it entirely.

The bottleneck isn’t the speed of coding, and what you’re doing here is actively worsening the situation.

9devon How I use LLMs to learn complex topics

I’m a bit too old for the YouTube generation (can’t stand the fast cuts and permanently agitated voices), but am absolutely guilty of reading a lot of pop-sci books. I feel like they do a better job at teaching stuff for the most part, but probably also make you fall into the trap of believing you actually learnt something…

On the other hand, I’m fine with not being an expert on topics outside of my domain, as long as I retain some basic knowledge and fun party facts. So there’s that.

9devon Show HN: Voice driven murder mystery, Interview AI suspects with your voice

The entire game hinges on the premise of being able to narrate and understand dynamic text on demand. Pre-recorded messages don’t work for this.

9devon Auto mode is now the default in Claude Code

What works in that case is adding a message saying "I authorise you to do $thing", which the classifier treats as explicit consent to do something and execute commands toward that goal.

9devon Auto mode is now the default in Claude Code

What an awful position to take. Tech used to be about becoming more accessible to people! Now we have a magical assistant to make computers do what you want with natural language, and your desire is gate keeping that so only programmers can use it to write software for themselves?

9devon The original URL for this prediction will no longer be available in 11 years (2011)

By I could, I meant that my phone has a persistent connection uninterruptedly long enough to be theoretically able to serve a website in the background. There are all sorts of (arbitrary, mind you) technical limitations, from the IPv6 address randomly changing as you roam across networks to iOS killing background processes - but those are software limitations.

The point was that mobile device doesn't equate spotty connectivity anymore for most people.

9devon The original URL for this prediction will no longer be available in 11 years (2011)

Is the p2p problem really still one in our world that is mobile first, but also has ubiquitous internet connectivity pretty much everywhere? I’m almost sure I could host my website from my phone and have at least 99.5% percent of uptime.

9devon The original URL for this prediction will no longer be available in 11 years (2011)

Someone should do a long bet on the continuous existence of longbets.com

9devon The original URL for this prediction will no longer be available in 11 years (2011)

That would not really pass the test in spirit, I think. Passing must be something that falls out as a consequence of being a good chatbot, not the other way around.

However such good conversational agents would still have benefits, and i would be curious how far the approach of building a model that’s nice to talk to can take you, as opposed to the current generation of all-knowledgable helpful assistants.

9devon The original URL for this prediction will no longer be available in 11 years (2011)

Most of the LLM tells we usually recognise are due to their heavy attunement to role playing a "helpful assistant" with a knack for correct typography.

If you used a smaller corpus, followed a different kind of fine tuning routine with rather normal conversation patterns, and wrote a different base prompt, I don’t see why it shouldn’t be possible to come up with a very particular personality, so to speak.

9devon Almost no skill required to cook a steak

It’s about experience. If you’re not into classical music, or jazz, or metal, it all sounds the same - how could anybody tell the difference? Only when you put in the work to listen to it a lot, different works, and actually pay attention, you will notice how your sense and understanding develops over time. Suddenly you’re able to make out differences, recognise patterns and repeated segments and structure where before there was none.

Or pick dancing: when you start out, you’re entirely occupied with placing your feet in the right spots at the right time. As you practice, your mind gets freed up and you’re able to make deliberate alterations, embellishments, larger figures. That seems impossible at the start.

9devon Tiny black holes may be exploding stars across the Milky Way

Many physicists have pointed out several flaws in the dark forest theory, which makes for a great read but doesn’t really carry over to the real universe 1:1.

9devon Stateless MCP has recaptured my interest

Kind of--I'm saying that RESTful APIs are just an abstract vocabulary to describe state transfer operations, whereas MCP describes a concrete set of such operations. REST sits on top of the HTTP protocol layer, MCP sits on top of REST (if we accept how the meaning of REST has been butchered, but you know what I mean.)

This is easy to see when you think in terms of a client: It is pretty much impossible to build an opinionated UI for any kind of RESTful API, while creating one for an MCP server is fairly self-explanatory: MCP servers offer a mandatory and complete runtime introspection endpoint (you can retrieve a listing of available tools/resources/prompts etc. along with their parameter and return type schemas). So that means clients have a way to exhaustively describe everything an MCP server is able to do with a vocabulary that carries over exactly to other servers - a tool is a tool everywhere.

9devon Waymo CEO explains why Tesla’s camera-only self-driving falls short

You're again thinking in terms of image sensors, which is not representative of what your body is able to do. For example, your brain processes signals from all your senses combined, which allows prioritisation of things that trigger multiple senses at once. It also applies pattern matching against millions of years of evolution's worth of information about danger.

But much more importantly, there is all sorts of post-processing happening that is so weird and complex and really a little decoupled from the raw visual information that it's futile to try and compare that to any digital operation you're thinking of in terms of cameras: If you were to look at a still captured from your eyes, you'd see an upside-down image that's only sharp in the very center of the image, blurry and desaturated pretty much everywhere else, grossly so in the corners, with several big black or blurry spots all over.

From that perspective, eyes are awful really. Yet still, your brain manages to turn the stereoscopic vision into a spatial representation, calculate the movement and keep track of other objects relative to you, fill in areas you can't actually see, detect things moving in your peripherals, highlight things of importance, and so much more.

9devon Stateless MCP has recaptured my interest

The spec is more than just an OpenAPI scheme; it’s a strict wire format with shared semantics for a huge range of features.

9devon Waymo CEO explains why Tesla’s camera-only self-driving falls short

Do you have any conception of how massively, incomprehensibly, extremely superior our sensory system is to any kind of camera-based computer vision we can make?

9devon The Warp Agent CLI

Well someone (like me) has to make the decision what their developers get to use, and that was in fact informed by subscription pricing vs. API cost.