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

I miss the old Claude Code

alexkras.com

Post reply on HN
Print view View original

21–26 of 26 posts

Re: I miss the old Claude Code

#21

Post by akras14 »

gorayobi wrote: post #14

I do too. In the past couple of months it's just doing too much. Before it was prompt -> model writes code -> done, I check the work. Now it's prompt -> model seems to read the entire codebase to 'get the full picture' -> takes forever to write code -> runs a whole bunch of things to verify the result. It seems like they have optimized the harness more and more for people who don't know what they are doing or what th…

Yes, author here. I was thinking about it more after the post.

I wonder if the harness has a bigger share of blame here than the model itself.

I seem to like using Opus 5 in the browsers a lot more... I.e. it was very helpful in fine-tuning my site after it went down due to HN traffic.

Re: I miss the old Claude Code

#23

Post by akras14 »

setnone wrote: post #4

it won me over at 4.6 and lost at 4.7 and the image 'How multiagent turf war runs ended' [0] might explain the shift in the behaviour [0] https://www.anthropic.com/research/multiagent-systems

I too keep setting Opus 4.6 manually in my daily work (when I am not using older Sonnet).

Re: I miss the old Claude Code

#25

Post by jeffybefffy519 »

packetlost wrote: post #17
jeffybefffy519 wrote: post #11

Earlier quoted context omitted.

Not commenting on speed, but it seems models and their harnesses have generally gotten much worse over time. My suspicion is that the "Frontier" labs really dont have a strong handle on good quality evals that equal expectations of their users, so they just churn out new models for marketing to sell the shit out of. Recently cancelled my Codex subscription because I cannot stand 5.6-sol/terra/luna. On top of that, Co…

As someone working in this space: unless they're a very big player with resources to create custom hardware, afaik there's no way to actually prevent the GPU host from seeing the request content. There's no "secure enclave" that sits between a CPU and the GPU, the decrypted payload must necessarily hit system RAM, as would the output. They are, at best, running in TEE VMs which are also pretty horribly broken and don…

You sure? These links are directly from tinfoil.sh's technology page: https://tinfoil.sh/technology

They use these technologies: - https://www.nvidia.com/en-us/data-center/solutions/confident... - https://www.amd.com/en/developer/sev.html - https://www.intel.com/content/www/us/en/products/details/pro...

Specifically, the intel page says "Intel® Xeon® 6 processors with Performance‑cores support Intel® TDX Connect technology, enabling confidential computing across the CPU and connected devices including GPUs, Smart NICs, and storage." which you are claiming do not exist....

Not trying to be an evangelist for them, but they seem to pretty open about their tech stack which is hugely differentiating compared to every other AI company....

Re: I miss the old Claude Code

#26

Post by lenkite »

Dathuil wrote: post #2

My first deep dive with Claude Code was in Jan of this year. I was working on rebuilding an old city builder game from the the 00s and was trying to parse the original images from the game files. There was plenty if info on the structure online so I said I'd let Claude give it a go. I was running one of the sonnets I think. After about 5 prompts I had a web interface that I could drag the files into and it presented…

Is this because of "model collapse" ? Newer AI models being trained on too much AI slop ?
Post reply on HN