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

Parallel AI agents are a game changer

morningcoffee.io

Post reply on HN
Print view View original

51–60 of 88 posts

Re: Parallel AI agents are a game changer

#51

Post by pvtmert »

ambicapter wrote: post #11

Obviously I'm an AI-tools skeptic, but this is hilarious: > 1. Prepare issues with sufficient context > Start by ensuring each GitHub issue contains enough context for agents to understand what needs to be built and how it integrates with the system. This might include details about feature behavior, file locations, database structure, or specific requirements such as displaying certain fields or handling edge cases.…

> This is just software engineering?!?

Indeed yes. Although most places shipping software in a "software development" and/or "programming" fashion for many years.

Many, many places certainly do not do the engineering part, even though resulting product is a software.

Re: Parallel AI agents are a game changer

#52

Post by rukuu001 »

ambicapter wrote: post #11

Obviously I'm an AI-tools skeptic, but this is hilarious: > 1. Prepare issues with sufficient context > Start by ensuring each GitHub issue contains enough context for agents to understand what needs to be built and how it integrates with the system. This might include details about feature behavior, file locations, database structure, or specific requirements such as displaying certain fields or handling edge cases.…

Yes, the ability to clearly and unambiguously communicate what's required works on both humans and machines.

Re: Parallel AI agents are a game changer

#53

Post by abound »

pvtmert wrote: post #42

Is it me or the post sounding ( showing! ) that they haven't tried the mentioned approach in real life. Because in real life, one agent tries to fix build issue with rm -rf node_modules and the other is already running a server (ie: npm server), conflicting with each other nearly all the time! . (even if it's not a destructive action, the second npm server will most likely to fail due to port-allocation conflicts!) M…

The post is about using GitHub's integrated Copilot tooling, where each issue gets its own instance presumably running in a sandbox. This sidesteps the issues you're talking about here.

Re: Parallel AI agents are a game changer

#54

Post by adriand »

I'm starting to think that the rather slow nature of Claude Code is a feature. In fact if they suddenly sped things up by 10X, I would want an option to slow it back down. Sometimes I am fine with it working unsupervised while I empty the dishwasher or take a shower, but a lot of the time I watch it work. Not only does this help me stop it from going down rabbit holes / chewing through all of my Opus usage cap, but I have a much better understanding of what it's built, in the same way I might if I was pair-programming with someone and they were driving.

The idea of having multiple instances working in parallel sounds like a nightmare to me. I find this technology works best when it is guided, ideally in real time.

Re: Parallel AI agents are a game changer

#55

Post by shiroyasha »

pvtmert wrote: post #42

Is it me or the post sounding ( showing! ) that they haven't tried the mentioned approach in real life. Because in real life, one agent tries to fix build issue with rm -rf node_modules and the other is already running a server (ie: npm server), conflicting with each other nearly all the time! . (even if it's not a destructive action, the second npm server will most likely to fail due to port-allocation conflicts!) M…

I don't claim to have lots of experience with this, I've been only doing it for a couple of weeks, but I do feel that some of your comments are disingenuous.

---

> Any other way multiple agents work in harmony in a single repo (filesystem/directory) at the same time is a pipe-dream with the current status of the MCP and agents.

Every agent runs in a separate VM on GitHub.

> Let alone being aware of each other (agents), they don't even have a proper locking mechanism.

Never claimed this. Feels like a strawman argument.

Re: Parallel AI agents are a game changer

#56

Post by pvtmert »

modarts wrote: post #33

It still amuses me how literally people took Kapathy's famous tweet around vibe coding https://x.com/karpathy/status/1886192184808149383 If people were to actually read beyond the first sentence, it would become clear very quickly that this was meant to be tongue in cheek.

Because most people have the context-window of 10 tokens, they do not read further than the first sentence (or two).

Re: Parallel AI agents are a game changer

#57

Post by wrs »

ambicapter wrote: post #11

Obviously I'm an AI-tools skeptic, but this is hilarious: > 1. Prepare issues with sufficient context > Start by ensuring each GitHub issue contains enough context for agents to understand what needs to be built and how it integrates with the system. This might include details about feature behavior, file locations, database structure, or specific requirements such as displaying certain fields or handling edge cases.…

Yeah, it’s funny, we may finally have a way to get developers to write documentation for other developers, it’s just that the other developers aren’t human!

Re: Parallel AI agents are a game changer

#58

Post by mmaunder »

hu3 wrote: post #50
mmaunder wrote: post #41

The author is lying. My team and I are heavy users of Claude code and other agents and it ain’t like this. You need to manage an AI coding agent carefully and course correct frequently. There are cases for parallel agents but they are tasks like parallel document fetches and summarization, and other tasks that don’t require supervision. The idea of having multiple parallel agents merge pull requests or resolve issues…

Your comment is disproportionately rude. Just because your team can't leverage multiple coding agents doesn't mean no one else can. And even if OP also can't, this is a good place to discuss possible problems and solutions for parallel development using coding agents. Please refrain from gatekeeping.

“With this approach, I can manage to have 10–20 pull requests open at once, each handled by a dedicated agent.”

A quote from the post. No, I think my post is calibrated quite well considering what OPs post does to our industry.

Re: Parallel AI agents are a game changer

#59

Post by skhameneh »

ambicapter wrote: post #11

Obviously I'm an AI-tools skeptic, but this is hilarious: > 1. Prepare issues with sufficient context > Start by ensuring each GitHub issue contains enough context for agents to understand what needs to be built and how it integrates with the system. This might include details about feature behavior, file locations, database structure, or specific requirements such as displaying certain fields or handling edge cases.…

> On the other hand, maybe I can convince people in my org to get better at software engineering by telling them its for the AI to work better.

Really good engineering practices are fundamental to get the most out of AI tooling. Convoluted documentation, code fragmentation, etc all pollute context when working with AI tools.

In my experience, just having one outdated word (especially if it's a library name) anywhere in code or documentation can create major ongoing headaches.

The worst part of it is trying to avoid negative assertions. When the AI tooling keeps trying to do "the wrong thing" it's sometimes a challenge to rephrase instructions for "the right thing" to frame a positive assertion.

Re: Parallel AI agents are a game changer

#60

Post by localhost »

zzzeek wrote: post #9

I use Claude every day. When I give it a program that does something straightforward in one file and it writes it from scratch, it does great. When I have it fix issues or add functionality to small to medium sized apps that have a mostly simple design, it does great. When I point it at codebases that are 20 years old and have a lot of indirection in their design due to years of hard lessons learned and a lot (like a…

one thing that i find works really well is to ask it to research things in the codebase and write a plan first. codex with gpt-5 is exceedingly good at doing this. then ask it to write a plan for what it would do with that information, i.e., i want you to research codebase for . then write a plan for how you would achieve given what you have learned.
Post reply on HN