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

Agents need control flow, not more prompts

bsuh.bearblog.dev

Post reply on HN
Print view View original

311–320 of 348 posts

Re: Agents need control flow, not more prompts

#311

Post by renticulous »

LikesPwsh wrote: post #271
krzyk wrote: post #258

Earlier quoted context omitted.

Could you elaborate what does "compiling orchestration prompt" mean?

Rather than telling the LLM "loop through these files", tell it "write a script to loop through these files", then hard-code that script somewhere.

First we gave LLMs access to bash commands. Now we give them access to customized commands which they can reuse. It's English language extending its claws into deterministic programming language. Now can we please have backtracking and dynamic programming like thinking loop built into English language or such orchestration prompts.

Re: Agents need control flow, not more prompts

#316

Post by rmaxdev »

827a wrote: post #100

1000% agree. I am increasingly hesitant to believe Anthropic's continual war drum of "build for the capabilities of future models, they'll get better". We've got a QA agent that needs to run through, say, 200 markdown files of requirements in a browser session. Its a cool system that has really helped improve our team's efficiency. For the longest time we tried everything to get a prompt like the following working: "…

The agent can do this one by one in an agentic loop, storing the progress and backlog in files

If nothing is stored in durable memory then the context window is going to get rotten

Re: Agents need control flow, not more prompts

#317

Post by hombre_fatal »

That this gets so much traction is an insight into the lack of process the average HNer is using while they say they can't get LLMs to do anything useful for them.

Turns out it really was just them expecting one-shots in Claude Code with "make no mistakes pls".

Something to keep in mind when listening to LLM discourse on HN.

Re: Agents need control flow, not more prompts

#318

Post by saltcured »

gman2093 wrote: post #291
jimbokun wrote: post #218

Earlier quoted context omitted.

So you have the model write the if statements and put itself out of a job.

Alternatively, and sometimes more cost-efficient: you can find a developer who can write bespoke if statements. There are dozens of us!

So, are we going to end up with a mechanical Turk that pretends it is an LLM but just farms out tasks to gig workers?

Re: Agents need control flow, not more prompts

#319

Post by tadfisher »

zapataband1 wrote: post #197
crsn wrote: post #156

Earlier quoted context omitted.

Our team at Agentforce recently open-sourced our solution to this and we've gotten very valuable feedback -- would love to hear from more of you about it: https://github.com/salesforce/agentscript

No you didn't "What we're not open sourcing (yet) is the runtime. "

If it actually takes off, expect a vibecoded runtime that everyone runs on their own systems.

Re: Agents need control flow, not more prompts

#320

Post by illwrks »

dnh44 wrote: post #150
illwrks wrote: post #68

I’ve been building a small ‘agent’ using copilot at work, partly a learning exercise as well as testing it in a small use case. My personal opinion is that AI and agents are being misrepresented… The amount of setup, guidance and testing that’s required to create smarter version of a form is insane. At the moment my small test is: Compressed instructions (to fit within the 8k limit) 9 different types of policies to g…

To be honest Copilot really stinks and is really far from the sharp edge of what is possible these days.

100%, have to use it for work though!
Post reply on HN