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

Parallel agents in Zed

zed.dev

Post reply on HN
Print view View original

81–90 of 178 posts

Re: Parallel agents in Zed

#81

Post by Frannky »

paulddraper wrote: post #66
Frannky wrote: post #61

I would love to unleash parallel agents, but I am still checking every single edit while enforcing minimal, stateless, modular code, and I have the AI check in with me before writing the next file. A lot of times, I find it has incredibly stupid ideas and tends to make the code very messy. I would love to figure out how to stop that from happening automatically. The upside of checking in on the code, though, is that…

> I would love to figure out how to stop that from happening automatically. AGENTS.md

I think the issue is deeper than prompts, agents.md, smart flows, etc. I think the problem is that LLMs are searchers, trained on preferring some results. So, if the dumb solution is there, and the smart solution is not there, they won't spit it out.

Re: Parallel agents in Zed

#82

Post by mswphd »

Frannky wrote: post #61

I would love to unleash parallel agents, but I am still checking every single edit while enforcing minimal, stateless, modular code, and I have the AI check in with me before writing the next file. A lot of times, I find it has incredibly stupid ideas and tends to make the code very messy. I would love to figure out how to stop that from happening automatically. The upside of checking in on the code, though, is that…

I'm expecting we'll likely end back up on agents making PRs, and having to review them. Either that or giving up on quality etc/dealing with very messy code. I've been trying various automated testing/linting/etc strategies, and they only work so well.

Re: Parallel agents in Zed

#83

Post by paulddraper »

jazzypants wrote: post #71
paulddraper wrote: post #66

Earlier quoted context omitted.

> I would love to figure out how to stop that from happening automatically. AGENTS.md

> AGENTS.md -- which will be ignored just often enough that you can never quite trust it.

And yet still less often than the average developer.

Re: Parallel agents in Zed

#84

Post by bigmadshoe »

jazzypants wrote: post #71
paulddraper wrote: post #66

Earlier quoted context omitted.

> I would love to figure out how to stop that from happening automatically. AGENTS.md

> AGENTS.md -- which will be ignored just often enough that you can never quite trust it.

Btw it may be obvious but afaik claude by default only reads CLAUDE.md and not AGENTS.md

Re: Parallel agents in Zed

#85

Post by Frannky »

mswphd wrote: post #82
Frannky wrote: post #61

I would love to unleash parallel agents, but I am still checking every single edit while enforcing minimal, stateless, modular code, and I have the AI check in with me before writing the next file. A lot of times, I find it has incredibly stupid ideas and tends to make the code very messy. I would love to figure out how to stop that from happening automatically. The upside of checking in on the code, though, is that…

I'm expecting we'll likely end back up on agents making PRs, and having to review them. Either that or giving up on quality etc/dealing with very messy code. I've been trying various automated testing/linting/etc strategies, and they only work so well.

That would be a nightmare. One thing is to review a PR generated by a human using AI and caring about the code; another is reviewing wild agents, especially when they make changes everywhere

Re: Parallel agents in Zed

#86

Post by sbysb »

jamie_ca wrote: post #7

I'm buying into this workflow more the more I use it, but the real gamechanger is (a) parallel threads in worktrees, with (b) enough lifecycle hooks to treat them similarly to spinning up a VM. Specifically for me that means that after I create a worktree I get some local config files copied over and Postgres duplicating my local dev and test databases so I can test in isolation, and then when I close out a worktree…

You should check out Ouijit [1] - I use it regularly for work and it's nice because it focuses on the environment that you want, and just gives you a shell that you can use any tooling in, as well as VM isolation per worktree if needed.

[1]: https://github.com/ouijit/ouijit

Re: Parallel agents in Zed

#87

Post by Petersipoi »

conqrr wrote: post #73

Love Zed for my personal projects at home with Openrouter. But I cant use it at work as I'm forced to use claude code at work and find the VSCode+claude code combo better. I know zed has claudecode integration, but I found it to be very weak compared to zedagent+claudemodel. Has this improved?

Do you mind sharing what your spend is like on Openrouter? I pay for the $200 Claude Code plan just so I never have to think about usage or feel like I'm paying per prompt. But I nearly max out my usage each week and I have no idea what that would cost me on Openrouter. I hear a lot of people are using MiniMax of Openrouter. Is that what you do?

Re: Parallel agents in Zed

#88

Post by kippinsula »

jamie_ca wrote: post #7

I'm buying into this workflow more the more I use it, but the real gamechanger is (a) parallel threads in worktrees, with (b) enough lifecycle hooks to treat them similarly to spinning up a VM. Specifically for me that means that after I create a worktree I get some local config files copied over and Postgres duplicating my local dev and test databases so I can test in isolation, and then when I close out a worktree…

tried parallel agents for a sprint and bounced off it. the worktree dance is fine, real blocker for us was test data isolation. scoped postgres schemas per branch worked, but reasoning about which agent broke teh shared migration when three of them touch it got old fast. we just run one agent at a time now and go for a walk.

Re: Parallel agents in Zed

#89

Post by ArielTM »

The worktree part is the easy half. Running parallel Claude Code subagents, the bottleneck is never "can they run without stomping each other's files." That's solved the moment each one has its own checkout.

The hard problem is architectural consistency. Agent A renames a type to X. Agent B, in a different worktree, independently renames the same type to Y because neither saw the other's decision. When you merge, neither worktree is "wrong" but the code is incoherent. You need either a shared decision log that every agent reads before starting, or an orchestrator that hands out scope narrow enough that no two agents can collide.

Zed's post is solving filesystem-level parallelism. The harder coordination problem is semantic, and that's where time savings from parallelization go to die.

Re: Parallel agents in Zed

#90

Post by zackangelo »

theappsecguy wrote: post #68
giancarlostoro wrote: post #50

Zed is probably the best text editor in the last 10 years. It has quirks, but it is insanely powerful and capable out of the box. I don't even bother trying to setup Neovim because of Zed. They let people for PRs for missing vim features for their vim emulation, and its insanely capable. I hope someday they get the funding they deserve, because it has insane potential. It's why I subscribe to their pay plan, even if…

I do wish they'd focus on closing the gap to Jetbrains by implementing the QOL features that are missing. I understand they have to do what VC wants to see, but this agentic stuff is so tiring.

I give them a try about twice a year. I write a lot of Rust which should be squarely in their wheelhouse.

This last time I was pleasantly surprised to find they mostly fixed their SSH remote editing support. But then it started truncating rustc inline error messages and I couldn’t figure out how to view the whole thing easily. When you’re just trying to get something done little bits like this can add up quickly. Punted back to Cursor for now.

Post reply on HN