4 ms·

That's kind of my benchmark for whether or not these models are useful. I've got a project that needs some extensive refactoring to get working again. Mostly up

That's kind of my benchmark for whether or not these models are useful. I've got a project that needs some extensive refactoring to get working again. Mostly upgrading packages, but also it will require updating the code to some new language semantics that didn't exist when it was written. So far, current AI models can make essentially zero progress on this task. I'll keep trying until they can!

tmpz22 1y ago
And IMO it has a long way to go. There is a lot of nuance when orchestrating dependencies that can cause subtle errors in an application that are not easily remedied. For example a lot of llms (I've seen it in Gemini 2.5, and Claude 3.7) will code non-existent methods in dynamic languages. While these runtime errors are often auto-fixable, sometimes they aren't, and breaking out of an agentic workflow to deep dive the problem is quite frustrating - if mostly because agentic coding entices us into being so lazy.
- jasonthorsness 1y ago
  The agents will definitely need a way to evaluate their work just as well as a human would - whether that's a full test suite, tests + directions on some manual verification as well, or whatever. If they can't use the same tools as a human would they'll never be able to improve things safely.
- mikepurvis 1y ago
  "... and breaking out of an agentic workflow to deep dive the problem is quite frustrating" Maybe that's the problem that needs solving then? The threshold doesn't have to be "bot capable of doing entire task end to end", like it could also be "bot does 90% of task, the worst and most boring part, human steps in at the end to help with the one bit that is more tricky". Or better yet, the bot is able to recognize its own limitations and proactively surface these instances, be like hey human I'm not sure what to do in this case; based on the docs I think it should be A or B, but I also feel like C should be possible yet I can't get any of them to work, what do you think? As humans, it's perfectly normal to put up a WIP PR and then solicit this type of feedback from our colleagues; why would a bot be any different?
  - dvfjsdhgfv 1y ago
    > Maybe that's the problem that needs solving then? The threshold doesn't have to be "bot capable of doing entire task end to end", like it could also be "bot does 90% of task, the worst and most boring part, human steps in at the end to help with the one bit that is more tricky". Still, the big short-term danger being you're left with code that seems to work well but has subtle bugs in it, and the long-term danger is that you're left with a codebase you're not familiar with.
    - mikepurvis 1y ago
      Being left with an unfamiliar codebase is always a concern and comes about through regular attrition, particularly if inadequate review is not in place or people are cycling in and out of the org too fast for proper knowledge transfer (so, cultural problems basically). If anything, I'd bet that agent-written code will get better review than average because the turn around time on fixes is fast and no one will sass you for nit-picking, so it's "worth it" to look closely and ensure it's done just the way you want.
- soperj 1y ago
  > if mostly because agentic coding entices us into being so lazy. Any coding I've done with Claude has been to ask it to build specific methods, if you don't understand what's actually happening, then you're building something that's unmaintainable. I feel like it's reducing typing and syntax errors, sometime it leads me down a wrong path.
  - weq 1y ago
    I can just imagine it now, you launch your AI coded first product and get a bug in production, and the only way the AI can fix the bug is to re-write and deploy the app with a different library. Your then proceed to show the changelog to the CCB for approval including explaining the fix to the client trying to explain its risk profile for their signoff. "Yeh, we solved the duplicate name appearing the table issue by moving databases engines and UI frameworks to ones more suited to the task"
yosito 1y ago
Personally, I don't believe AI is ever going to get to that level. I'd love to be proven wrong, but I really don't believe that an LLM is the right tool for a job that requires novel thinking about out of the ordinary problems like all the weird edge cases and poor documentation that comes up when trying to upgrade old software.
- 9dev 1y ago
  Actually, I think the opposite: Upgrading a project that needs dependency updates to new major versions—let’s say Zod 4, or Tailwind 3—requires reading the upgrade guides and documentation, and transferring that into the project. In other words, transforming text. It’s thankless, stupid toil. I’m very confident I will not be doing this much more often in my career.
  - mikepurvis 1y ago
    Absolutely, this should be exactly the kind of task a bot should be perfect for. There's no abstraction, no design work, no refactoring, no consideration of stakeholders, just finding instances of whatever is old and busted and changing it for the new hotness.
    - dvfjsdhgfv 1y ago
      So that was my conviction, too. However, in my tests it seems like upgrading to a version a model hasn't seen is for some reason problematic, in spite of giving it the complete docs, examples of new API usage etc. This happens even with small snippets, even though they can deal with large code fragments with older APIs they are very "familiar" with.
      - mikepurvis 1y ago
        Okay so less of a "this isn't going to work at all" and more just not ready for prime-time yet.
    - maoberlehner 1y ago
mewpmewp2 1y ago
I think this type of thing needs agent which has access to the documentation to read about nuances of the language and package versions, definitely a way to investigate types, interfaces. Problem is that training data has so much mixed data it can easily confuse the AI to mix up versions, APIs etc.

tmpz22 1y ago

And IMO it has a long way to go. There is a lot of nuance when orchestrating dependencies that can cause subtle errors in an application that are not easily remedied. For example a lot of llms (I've seen it in Gemini 2.5, and Claude 3.7) will code non-existent methods in dynamic languages. While these runtime errors are often auto-fixable, sometimes they aren't, and breaking out of an agentic workflow to deep dive the problem is quite frustrating - if mostly because agentic coding entices us into being so lazy.

jasonthorsness 1y ago
The agents will definitely need a way to evaluate their work just as well as a human would - whether that's a full test suite, tests + directions on some manual verification as well, or whatever. If they can't use the same tools as a human would they'll never be able to improve things safely.
mikepurvis 1y ago
"... and breaking out of an agentic workflow to deep dive the problem is quite frustrating" Maybe that's the problem that needs solving then? The threshold doesn't have to be "bot capable of doing entire task end to end", like it could also be "bot does 90% of task, the worst and most boring part, human steps in at the end to help with the one bit that is more tricky". Or better yet, the bot is able to recognize its own limitations and proactively surface these instances, be like hey human I'm not sure what to do in this case; based on the docs I think it should be A or B, but I also feel like C should be possible yet I can't get any of them to work, what do you think? As humans, it's perfectly normal to put up a WIP PR and then solicit this type of feedback from our colleagues; why would a bot be any different?
- dvfjsdhgfv 1y ago
  > Maybe that's the problem that needs solving then? The threshold doesn't have to be "bot capable of doing entire task end to end", like it could also be "bot does 90% of task, the worst and most boring part, human steps in at the end to help with the one bit that is more tricky". Still, the big short-term danger being you're left with code that seems to work well but has subtle bugs in it, and the long-term danger is that you're left with a codebase you're not familiar with.
  - mikepurvis 1y ago
    Being left with an unfamiliar codebase is always a concern and comes about through regular attrition, particularly if inadequate review is not in place or people are cycling in and out of the org too fast for proper knowledge transfer (so, cultural problems basically). If anything, I'd bet that agent-written code will get better review than average because the turn around time on fixes is fast and no one will sass you for nit-picking, so it's "worth it" to look closely and ensure it's done just the way you want.
soperj 1y ago
> if mostly because agentic coding entices us into being so lazy. Any coding I've done with Claude has been to ask it to build specific methods, if you don't understand what's actually happening, then you're building something that's unmaintainable. I feel like it's reducing typing and syntax errors, sometime it leads me down a wrong path.
- weq 1y ago
  I can just imagine it now, you launch your AI coded first product and get a bug in production, and the only way the AI can fix the bug is to re-write and deploy the app with a different library. Your then proceed to show the changelog to the CCB for approval including explaining the fix to the client trying to explain its risk profile for their signoff. "Yeh, we solved the duplicate name appearing the table issue by moving databases engines and UI frameworks to ones more suited to the task"

yosito 1y ago

Personally, I don't believe AI is ever going to get to that level. I'd love to be proven wrong, but I really don't believe that an LLM is the right tool for a job that requires novel thinking about out of the ordinary problems like all the weird edge cases and poor documentation that comes up when trying to upgrade old software.

9dev 1y ago
Actually, I think the opposite: Upgrading a project that needs dependency updates to new major versions—let’s say Zod 4, or Tailwind 3—requires reading the upgrade guides and documentation, and transferring that into the project. In other words, transforming text. It’s thankless, stupid toil. I’m very confident I will not be doing this much more often in my career.
- mikepurvis 1y ago
  Absolutely, this should be exactly the kind of task a bot should be perfect for. There's no abstraction, no design work, no refactoring, no consideration of stakeholders, just finding instances of whatever is old and busted and changing it for the new hotness.
  - dvfjsdhgfv 1y ago
    So that was my conviction, too. However, in my tests it seems like upgrading to a version a model hasn't seen is for some reason problematic, in spite of giving it the complete docs, examples of new API usage etc. This happens even with small snippets, even though they can deal with large code fragments with older APIs they are very "familiar" with.
    - mikepurvis 1y ago
      Okay so less of a "this isn't going to work at all" and more just not ready for prime-time yet.
  - maoberlehner 1y ago

mewpmewp2 1y ago

I think this type of thing needs agent which has access to the documentation to read about nuances of the language and package versions, definitely a way to investigate types, interfaces. Problem is that training data has so much mixed data it can easily confuse the AI to mix up versions, APIs etc.