Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Sonnet 4.5

anthropic.com

Post reply on HN
Print view View original

521–530 of 819 posts

Re: Claude Sonnet 4.5

#521

Post by manofmanysmiles »

I haven't shouted into the void for a while. Today is as good a day as any other to do so.

I feel extremely disempowered that these coding sessions are effectively black box, and non-reproducible. It feels like I am coding with nothing but hopes and dreams, and the connection between my will and the patterns of energy is so tenuous I almost don't feel like touching a computer again.

A lack of determinism comes from many places, but primarily: 1) The models change 2) The models are not deterministic 3) The history of tool use and chat input is not availabler as a first class artifact for use.

I would love to see a tool that logs the full history of all agents that sculpt a codebase, including the inputs to tools, tool versions and any other sources of enetropy. Logging the seed into the RNGs that trigger LLM output would be the final piece that would give me confidence to consider using these tools seriously.

I write this now after what I am calling "AI disillusionment", a feel where I feel so disconnected from my codebase I'd rather just delete it than continue.

Having a set of breadcrumbs would give me at least a modicum of confidence that the work was reproducible and no the product of some modern ghost, completely detached from my will.

Of course this would require actually owning the full LLM.

Re: Claude Sonnet 4.5

#522

Post by tymscar »

tclancy wrote: post #479
Aflynn50 wrote: post #374

When I see how much the latest models are capable of it makes me feel depressed. As well as potentially ruining my career in the next few years, its turning all the minutiae and specifics of writing clean code, that I've worked hard to learn over the past years, into irrelivent details. All the specifics I thought were so important are just implementation details of the prompt. Maybe I've got a fairly backwards view…

Eh, I’ve gotten over that. I’ve been using Claude recently on a personal project for a friend who wanted to take a known export file format and turn it into a list of good households for local political candidates to hit when knocking on doors. And I did that. But it’s been a while since I used pandas and numpy so I told Claude to swap out my loops for efficient code. And he did. Then, just for fun, I said, “Hey, sin…

I 100% agree with you, but to play devils advocate, what would stop an LLM for telling you all about k means clustering and openstreetmap and everything when you ask about an efficient way to cluster deliveries on a map?

Re: Claude Sonnet 4.5

#523

Post by johnfn »

manofmanysmiles wrote: post #521

I haven't shouted into the void for a while. Today is as good a day as any other to do so. I feel extremely disempowered that these coding sessions are effectively black box, and non-reproducible. It feels like I am coding with nothing but hopes and dreams, and the connection between my will and the patterns of energy is so tenuous I almost don't feel like touching a computer again. A lack of determinism comes from m…

If you care about this so much why don't you use one of the open source OpenAI models? They're pretty good and give you the guarantees you want.

Re: Claude Sonnet 4.5

#527

Post by stingraycharles »

Implicated wrote: post #317
iagooar wrote: post #296

Earlier quoted context omitted.

I think that is an interesting observation and I generally agree. Your point about prompting quality is very valid and for larger features I always use PRDs that are 5-20x the prompt. The thing is my "experiment" is one that represents a fairly common use case: this feature is actually pretty small and embeds into an pre-existing UI structure - in a larger codebase. GPT-5-Codex allows me to write a pretty quick & dir…

> The thing is my "experiment" is one that represents a fairly common use case Valid as well. I guess I'm just nitpicking based on how much I see people saying these models aren't useful combined with seeing this example, triggered my "you're doing it wrong" mode :D > GPT-5-Codex allows me to write a pretty quick & dirty prompt, yet still get VERY good results. I have a reputation with family and co-workers of being…

Yeah, as someone who has been using Claude Code for about 4 months now, I’ve adopted a “be super specific by default”-workflow. It works very well.

I typically use zen-mcp-server’s planning mode to scope out these tasks, refine and iterate on a plan, clear context, and then trigger the implementation.

There’s no way I would have considered “implement fuzzy search” a small feature request. I’m also paranoid about introducing technical debt / crappy code, as in my experience is the #1 reason that LLMs typically work well for new projects but start to degrade after a while: there’s just a lot of spaghetti and debt built up over time.

Re: Claude Sonnet 4.5

#528

Post by wyre »

paool wrote: post #469
pton_xd wrote: post #287

Earlier quoted context omitted.

> I bet if I were in your shoes and looking to write a prompt to start a task of a similar type that my prompt would have been 5 to 20x the length of yours Why would you need such extensive prompting just to get the model to not re-implement authentication logic, for example? It already has access to all of the existing code, shouldn't it just take advantage of what's already there? A 20x longer prompt doesn't sound…

> shouldn't it just take advantage of what's already there? It's not a good idea to have any coding agent put unnecessary amounts of lines into the context window in order to understand your code base. Performance of all llms drop drastically when the context window is filled or full. The purpose of being more specific with your prompts is that you spend a little bit more tokens up front to make the task a lot more e…

I'm not sure exactly what you mean by the vibe coding experience non-coders want, but if it's one-shotting a buildable codebase off of an unspecific prompt, the major breakthrough would have to be brain-computer interfaces so the agent can literally read the user's mind.

If that same person approached a software development company with the same prompt without following up with any other details, they won't get good code back, either. You're not saying it, but this idea that in the future you can tell a computer something like "create photoshop" and get what your expecting is an unrealistic dream that would need mind-reading or a major breakthrough and paradigm shift in understanding and interpreting language.

Re: Claude Sonnet 4.5

#529

Post by Galaco »

If you pause your subscription, Claude.ai breaks. I paused my subscription, and my account immediately transitioned to free. It has removed my invoice history, and attempts to upgrade again fail with an internal error. Their chatbot is telling me to navigate to UI elements that don't exist, and free users do not have the option of human support.

So I'm stuck; my sub is paused, and I cannot either cancel, or unpause and cannot speak to a human to solve this because the pause process took away all possibility of human interaction.

This is the future we live in.

Re: Claude Sonnet 4.5

#530

Post by shanecp »

Bjorkbat wrote: post #169

> Practically speaking, we’ve observed it maintaining focus for more than 30 hours on complex, multi-step tasks. Really curious about this since people keep bringing it up on Twitter. They mention it pretty much off-handedly in their press release and doesn't show up at all in their system card. It's only through an article on The Verge that we get more context. Apparently they told it to build a Slack clone and left…

What they don't mention is all the tooling, MCPs and other stuff they've added to make this work. It's not 30 hours out of the box. It's probably heavily guard-railed, with a lot of validated plans, checklists and verification points they can check. It's similar to 'lab conditions', you won't get that output in real-world situations.
Post reply on HN