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

Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

news.ycombinator.com

Post reply on HN
Print view View original

181–190 of 196 posts

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#181

Post by bhaktatejas922 »

helsinki wrote: post #159
bhaktatejas922 wrote: post #78

Earlier quoted context omitted.

yeah same. I feel like Opus tends to be slightly more sycophancy leaning on technical topics

Interesting. I use Opus exclusively (like $1000/day in tokens) via Claude Code. Do you really think Sonnet is better for programming? I’m not sure I agree, though I’d love to save $900/day by taking you up on it.

Im not sure that it's "better". I still use Opus and it's better at coding but needs more steering to be less of a "Yes you're EXACTLY right" every time i suggest a new solution path. Purely anecdotal though

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#182

Post by thegeomaster »

conartist6 wrote: post #173
thegeomaster wrote: post #164

Earlier quoted context omitted.

LLMs will generate unpredictable, very humanlike code edits in this form. They might use comments like "same function as above", "rest of the function with similar changes", "function ABC is no longer needed", "function ABC same as above", etc. Your code edit model must resolve all of these, or flag an error in case of too much ambiguity. I would think a classical algorithm would have a lot of trouble differentiating…

I agree that sorting something that messy can't easily be done with a heuristic expert system, but I'm registering my concern that the whole approach is predicated on first making a huge mess with one LLM then cleaning up the mess with another. The classical approach is more like "change the definition of the problem until you don't need to make a big mess in the first place"

This might be a controversial take, but this approach is just plain old engineering applied to LLMs.

Instead of making an LLM perform both an accurate code edit AND follow a strict output schema, you split that into two problems: accurate code edit with a lax output schema, then application of that lax output schema to the original file. You can then use different models for the two tasks, reducing your probability of failure.

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#183

Post by handfuloflight »

klank wrote: post #172
handfuloflight wrote: post #171

Earlier quoted context omitted.

Next time, I'll wrap it up in . Truly ironic the AI readily detected what I said as sarcastic. Without context. https://claude.ai/share/7d14287d-c066-4927-8942-8eb8dd8d7e7f

Ah, thanks for the explination. I actually was confused a bit. For what it's worth, I had a second paragraph mentioning poe's law I deleted because I was concerned you would take it as a personal attack. I should have left it in, knowing you were sarcastic I think you'd have appreciated me being confused about whether you were being satirical or not.

That would have been a perfect opportunity for me to finally internalize Poe's law.

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#184

Post by Mo3 »

godot wrote: post #136
ashwindharne wrote: post #31

Earlier quoted context omitted.

I do find that having inference happen ~50% faster is much more valuable to my workflow than a single digit accuracy increase. If I'm going to have to check that the changes are correct anyways, getting more iterations in faster feels much better than incremental accuracy. There's definitely a tipping point though. If the accuracy gains are so high that I can check its work less carefully or less often, the benefits…

I've been using Cursor pretty extensively in the past few months and I use it to code pretty hard problems sometimes, and a while ago when the options were between claude 3.5 sonnet vs gemini 2.5 pro, there was such a significant difference in quality that claude 3.5 often straight up failed -- the code it wrote woudln't work, even after retrying over and over again, and gemini 2.5 pro often was able to solve it corr…

Yup, same. My Google API costs were way too high. Sonnet and Opus 4 are much better now so they take care of most of my "easier" tasks. Gemini 2.5 Pro is still somehow better for larger scopes so I have it do all the pre-planning and larger tasks

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#185

Post by klank »

handfuloflight wrote: post #183
klank wrote: post #172

Earlier quoted context omitted.

Ah, thanks for the explination. I actually was confused a bit. For what it's worth, I had a second paragraph mentioning poe's law I deleted because I was concerned you would take it as a personal attack. I should have left it in, knowing you were sarcastic I think you'd have appreciated me being confused about whether you were being satirical or not.

That would have been a perfect opportunity for me to finally internalize Poe's law.

Haha. You got me. I couldn't tell. I really couldn't tell.

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#186

Post by bhaktatejas922 »

dsp_person wrote: post #151

2. Sometimes watching Claude do inline edits I cringe watching it delete almost the entire file line by line. I think when early in exploring a problem it's best to torch the previous version and fully re-write than to assume incremental.

Yeah, incremental updates are something models struggle with, which is what we're trying to solve.

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#187

Post by conartist6 »

thegeomaster wrote: post #182
conartist6 wrote: post #173

Earlier quoted context omitted.

I agree that sorting something that messy can't easily be done with a heuristic expert system, but I'm registering my concern that the whole approach is predicated on first making a huge mess with one LLM then cleaning up the mess with another. The classical approach is more like "change the definition of the problem until you don't need to make a big mess in the first place"

This might be a controversial take, but this approach is just plain old engineering applied to LLMs. Instead of making an LLM perform both an accurate code edit AND follow a strict output schema, you split that into two problems: accurate code edit with a lax output schema, then application of that lax output schema to the original file. You can then use different models for the two tasks, reducing your probability o…

Yeah, it's just that I'm building things to last the next 50 years. Of course you can build a skyscraper on a foundation of wet noodles and it would still be (impressive) engineering, but the structure would not stand the test of time.

So yeah, it's a clever approach, and useful even, but there's no way I can see that that such a noodl-y hack could become the bedrock for all other infrastructure

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#188

Post by helsinki »

stoken wrote: post #165
helsinki wrote: post #159

Earlier quoted context omitted.

Interesting. I use Opus exclusively (like $1000/day in tokens) via Claude Code. Do you really think Sonnet is better for programming? I’m not sure I agree, though I’d love to save $900/day by taking you up on it.

Genuine: how? I assume you're using something like cc-usage to get that value. $1k/day is tons. Would genuinely love to know how you're managing to keep the inference burning through that much a day, as I'd love to do the same, but even with 2-4 simultaneous sessions running fairly continuously, mostly on Opus for 10-12 hours a day, I get maybe $500/day. What's your workflow rig/setup look like to get you to that $1k…

I use Vertex and work at a hedge fund. I just spam Claude Code Opus all day long. There’s not much to it, other than I sit at a chair for 12-16 hours and spam poor (actually, rich) Claude. I don’t use the cc usage too - I just look at my GCP bill :(

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#189

Post by bpt3 »

dingnuts wrote: post #112
sitkack wrote: post #111

Earlier quoted context omitted.

Flow state has been redefined now that we are all using Claude Code. If I can stay focused on tests, reviewing code, etc while CC is doing its thing, we are good. The kloc/s doesn't matter as much.

if LLMs are ever able to write the kind of code I write for work, I'm going to move to management. spending 100% of my time reviewing AI slop and writing tests is the opposite of what I want. I want to define behavior quickly and have AI do the boring parts; you're letting the computer do the fun bit and spending your entire life doing the shit part, and paying for the privilege. fuck. THAT.

I might put this on a plaque.

I realize this sounds harsh, but I assume anyone who is pushing for developers to basically take on all the shit work of a tech lead stuck managing a bunch of incompetent developers is not an actual developer, and is either an incompetent one who hopes LLMs will cover for them or someone looking to reduce their dependency on developers.

Fortunately for me, I think we'll be well into the Matrix before my job can be done adequately by AI so I have the luxury of using it as a tool here and there where it makes sense rather than spending most of my time trying to avoid the damage a firehose of hallucinations will do to my codebase.

Re: Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec

#190

Post by bpt3 »

ipaddr wrote: post #128
dingnuts wrote: post #112

Earlier quoted context omitted.

if LLMs are ever able to write the kind of code I write for work, I'm going to move to management. spending 100% of my time reviewing AI slop and writing tests is the opposite of what I want. I want to define behavior quickly and have AI do the boring parts; you're letting the computer do the fun bit and spending your entire life doing the shit part, and paying for the privilege. fuck. THAT.

We have it backwards. Claude should be reviewing work and writing tests.

No one sane would trust an LLM with that task, which is how we know it's not ready for production use yet.
Post reply on HN