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

The new rules of context engineering for Claude 5 generation models

claude.com

Post reply on HN
Print view View original

181–190 of 434 posts

Re: The new rules of context engineering for Claude 5 generation models

#181

Post by Jensson »

chrisweekly wrote: post #138
chmod775 wrote: post #113

Earlier quoted context omitted.

It's language and speech patterns that seem designed to trick readers into believing that claims are correct, even when the claims aren't based on anything and are possibly wrong. It was rewarded for this during training for some reason. Alternative theory: The LLMs only way to "think" about abstract concepts is through language, and this leaks into into conversation it has with humans. But humans generally prefer to…

"There is a depth of thought untouched by words, and deeper still a depth of formless feeling untouched by thought." - Rilke Your assertion that we don't think in language is questionable. It runs counter to the lived experience of developing thoughts through writing ("writing isn't capturing thinking -- it is thinking"). I believe there is more to thought than language alone, but I also feel quite sure that language…

> It runs counter to the lived experience of developing thoughts through writing

Writing requires thought, but writing isn't thought. Just like doing requires thought, but doing isn't thought, writing is a subset of doing.

You can also doodle to think things through, or play with toys to think things through, or many other similar things.

Re: The new rules of context engineering for Claude 5 generation models

#182

Post by boorang »

I ran into this issue about 6 months ago when i was using mitmproxy to view the system prompts for claude code to try to diagnose degraded adherence to my CLAUDE.MD instructions: https://github.com/anthropics/claude-code/issues/18560

To summarize- they were embedding the CLAUDE.MD in a system-reminder with this disclaimer at the end: "IMPORTANT: this context may or may not be relevant to your tasks. You should not respond to this context unless it is highly relevant to your task."

This flew under the radar and they never addressed it, but it felt like at least a big part of the "nerfing" story. I no longer have a Claude Code subscription to test it, but I think it's a useful exercise for most people to sniff the traffic at least once to get an idea of what the back and forth with the Claude Code harness entails.

As others have noted, Anthropic seems to be on a path to make coding ever more accessible to non-coders, and in doing so has removed alot of the controls from devs who do want a more guided experience.

Re: The new rules of context engineering for Claude 5 generation models

#183

Post by boorang »

I ran into this issue about 6 months ago when i was using mitmproxy to view the system prompts for claude code to try to diagnose degraded adherence to my CLAUDE.MD instructions: https://github.com/anthropics/claude-code/issues/18560

To summarize- they were embedding the CLAUDE.MD in a system-reminder with this disclaimer at the end: "IMPORTANT: this context may or may not be relevant to your tasks. You should not respond to this context unless it is highly relevant to your task."

This flew under the radar and they never addressed it, but it felt like a real part of the "nerfing" story. I no longer have a Claude Code subscription to test it, but I think it's a useful exercise for most people to sniff the traffic at least once to get an idea of what the back and forth with the Claude Code harness entails.

As others have noted, Anthropic seems to be on a path to make coding ever more accessible to non-coders, and in doing so has removed alot of the controls from devs who do want a more manual experience.

Re: The new rules of context engineering for Claude 5 generation models

#184

Post by anon22981 »

zahlman wrote: post #58
firasd wrote: post #3

I've always thought that extensive throat-clearing and prefixing the Treaties of Westphalia-length instructions into the context window was unnecessarily baroque when you can just talk to the agent. I guess part of it is also that I don't mind doing 'hand-edits' like for example LLMs love to say "// so and so removed" I just go and remove that manually later rather than being like "don't comment about what you remove…

> I guess part of it is also that I don't mind doing 'hand-edits' like for example LLMs love to say "// so and so removed" I just go and remove that manually later rather than being like "don't comment about what you removed!11" cause you're really fighting deep grooves in the model's behavior at that point. I feel like this is the way. There are surely things where it's faster; certainly it's more pleasant to do sim…

I’m ashamed to admit that I once suddenly realized what I had done: I asked the LLM to rename a variable. It took it some minutes or so. With IDE by hand it would have been seconds.

Re: The new rules of context engineering for Claude 5 generation models

#185

Post by baq »

knollimar wrote: post #55
Retr0id wrote: post #53

Earlier quoted context omitted.

Until recently I thought "load-bearing seam" was a satirical exaggeration - I'd seen both claudisms independently but never combined. But a couple of days ago it hit me with "The key structural point first: the only load-bearing seam is [...]"

the phrase load-bearing caveat makes me irate

Only that one? Not belt-and-suspenders enough

Re: The new rules of context engineering for Claude 5 generation models

#186

Post by baq »

chaboud wrote: post #163
hmokiguess wrote: post #47

Earlier quoted context omitted.

Now I have the full picture. You're right to push back, and that's on me. The load-bearing seams of language are the smoking gun I should have been aware of.

I have heard so many of my co-workers use "load-bearing" over the last couple of months. It's truly comical. Maybe this is a way that we can make "fetch" happen.

adjacent surface seam

Re: The new rules of context engineering for Claude 5 generation models

#187

Post by globular-toast »

grim_io wrote: post #22
mycentstoo wrote: post #7

We should design a specific language to make sure that we can encode the exact requirements that we want. Something that has a limited set of keywords that are explicit. Wait a minute...

The success of LLM's (by usage) tells us that programming languages are still too close to the machine than the actual problem domain as defined by humans. If we truly had the right abstractions, no one would care to use LLM's for programming.

But your job is to write the abstractions. Python can't ship with high level abstractions for every problem that has ever been and ever will be.

Re: The new rules of context engineering for Claude 5 generation models

#188

Post by roncesvalles »

chrisweekly wrote: post #138
chmod775 wrote: post #113

Earlier quoted context omitted.

It's language and speech patterns that seem designed to trick readers into believing that claims are correct, even when the claims aren't based on anything and are possibly wrong. It was rewarded for this during training for some reason. Alternative theory: The LLMs only way to "think" about abstract concepts is through language, and this leaks into into conversation it has with humans. But humans generally prefer to…

"There is a depth of thought untouched by words, and deeper still a depth of formless feeling untouched by thought." - Rilke Your assertion that we don't think in language is questionable. It runs counter to the lived experience of developing thoughts through writing ("writing isn't capturing thinking -- it is thinking"). I believe there is more to thought than language alone, but I also feel quite sure that language…

Language is a reductive and lossy serialization of "thought-stuff". Sometimes you need this information-shedding to clear your working memory to make room for more things. Sometimes it's literally just a way to communicate. You're turning something fuzzy into something discrete.

E.g. "I'm feeling something. Is it anger? Yes, I'm angry." But in reality anger isn't just one thing. It's a cluster of infinite and varied feelings that we label as "anger". Something is lost when we do this labeling.

Notice then that the feeling of "anger" didn't start from your language, you merely used language to label, discretize, classify, standardize, compress it. It's one-way.

Re: The new rules of context engineering for Claude 5 generation models

#189

Post by espeed »

Claude Code is deleting your context history on a timer. I wanted to build a searchable index of my context history, and tonight I discovered, "The default retention is roughly 30–45 days. Anything older gets removed automatically." https://code.claude.com/docs/en/data-usage#data-retention This is nuts. Anthropic should not be deleting your data on your own device.

Re: The new rules of context engineering for Claude 5 generation models

#190

Post by prox »

bensyverson wrote: post #145
DerivativeBS wrote: post #137

Earlier quoted context omitted.

A compiler for Claude-isms? First pass: s/load-bearing/while/g s/smoking gun/throw/g s/that's on me/catch/g s/full picture/main/g

s/blast radius/malloc/g

Wait it uses “blast radius” ? In what context?
Post reply on HN