[dead]
Agents need control flow, not more prompts
301–310 of 348 posts
Re: Agents need control flow, not more prompts
#302Earlier quoted context omitted.
100% agreed. use the non-deterministic thing that is right 90% of the time to generate a deterministic thing that is right 100% of the time. one of the key things I add to my prompts is: - Please consult me when you encounter any ambiguous edge cases Attaching the AI to production to directly do things with API calls is bad. For me the only use case where the app should do any AI stuff is with reading/categorizing/et…
> Please consult me when you encounter any ambiguous edge cases Why not check the logprobs of the output and take action when the prob of the first and second most likely token is too similar? (or below a certain threshold?
I think you're getting abstraction layers mixed up, prediction uncertainty and logical uncertainty aren't the same. In a reasoning model, it's entirely possible that there's only one likely continuation and it says something like "This edge case is ambiguous, but what the user most likely meant is X".
Re: Agents need control flow, not more prompts
#3031000% 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: "…
I m running into similar issues, more and more i’m removing complexity from the agent to the (Go) logic in order to make it more deterministic.
To be more precise; everything is prepared in the form of files instead of letting the subagents making api/cli calls. And still - sometimes (even with enough context) the main agent takes strange turns.
Re: Agents need control flow, not more prompts
#304[flagged]
Re: Agents need control flow, not more prompts
#305[flagged]
Re: Agents need control flow, not more prompts
#306Earlier quoted context omitted.
> Slash commands, for instance, are a misfeature. I should never have to wait for the chatbot finish a turn so that I can check on the status of my context window or how much money I've spent this session. Control should be orthogonal to the chat loop. I get what you're trying to say but in practice architecting what you propose is considerably more difficult. Why not build it and try to get hired by one of the bigco…
In what way would it be more complicated? This is pretty basic concurrent programming, we routinely have much much more complex concurrent designs.. Hell, a telegram bot can handle that just fine.
Yeah, basic concurrent programming is not more complicated than basic linear programming
Re: Agents need control flow, not more prompts
#307Seems more and more people are coming to the same realization:
https://community.safebots.ai/t/prominent-people-come-to-the...
Re: Agents need control flow, not more prompts
#308Relying on prompt engineering for logic is incredibly fragile. Explicit state machines and programmatic routing provide the predictability that complex agents actually require.
Re: Agents need control flow, not more prompts
#3091000% 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: "…
I saw a major uplift in performance after I combined tools like apply_patch with check_compilation & run_unit_tests. I still call the tool "apply_patch", but it now returns additional information about the build & tests if the patch succeeds. The agent went from ~80% success rate to what seems to be deterministic (so far). I don't bother to describe the compilation and unit testing processes in my prompts anymore. Al…
I like this - I think you're not too far off of what's popular these days though. I think similar functionality can be achieved by using the "hook" functionality in claude code / codex.
Re: Agents need control flow, not more prompts
#3101000% 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: "…
I almost always use orchestration tooling nowadays. cc itself feels too basic, even with things like superpowers.