Backpressure is all you need
lucasfcosta.com
Backpressure is all you need
1–10 of 133 posts
Re: Backpressure is all you need
#2Interesting ideas for generalizing goals to reduce human labor in human agent interactions. That said, maybe it is better to set up customized skills and infrastructure for large projects? At our early stage of trying to capture value of agentic systems, the good ideas in this article might be premature optimization.
Re: Backpressure is all you need
#3> It should also reduce the number of low-quality PRs your teammates have to review for details the agent should have caught itself.
Oh boy.
Re: Backpressure is all you need
#4I’m willing to be wrong but this industry-wide emphasis on AI creative/coding workflows seems way over-engineered.
Ime successful creative execution looks like micro-iterations where each output informs the next creative move.
I can build something incredibly fast from essentially caveman grunt instructions through an LLM harness, iterating as I go.
Optimizing for feeding a huge plan to an agent sounds to me like a net waste of time. And looking over the shoulder of industry peers trying to do this, I don’t see their outputs or throughput some remarkable improvement over what I can produce with minimal fanfare usage.
Re: Backpressure is all you need
#5I’m willing to be wrong but this industry-wide emphasis on AI creative/coding workflows seems way over-engineered. Ime successful creative execution looks like micro-iterations where each output informs the next creative move. I can build something incredibly fast from essentially caveman grunt instructions through an LLM harness, iterating as I go. Optimizing for feeding a huge plan to an agent sounds to me like a n…
Yeah it's wild watching so many people decide waterfall is great all of a sudden.
Re: Backpressure is all you need
#6Oh this is 101. Anyone not doing this? If not do it now!
Re: Backpressure is all you need
#7I’m willing to be wrong but this industry-wide emphasis on AI creative/coding workflows seems way over-engineered. Ime successful creative execution looks like micro-iterations where each output informs the next creative move. I can build something incredibly fast from essentially caveman grunt instructions through an LLM harness, iterating as I go. Optimizing for feeding a huge plan to an agent sounds to me like a n…
The trick is to have 5 of those huge plans running in parallel.
Re: Backpressure is all you need
#8[flagged]
Re: Backpressure is all you need
#9"In this post, I’ll cover a third, not-so-obvious approach: building ways for the agent to validate more of its own work before a human has to step in. "
this has been an obvious thing to do since at least January (since Geoffrey Huntley published "everything is a ralph loop"), and this is how I've been working: build enough orchestration tooling to be able to automate everything: development container bringup, building it, running the unit tests, doing integration testing, and using the software as eventually an end user. then to iterate set performance goals on an already solid basis so the automated agent ("gym") can go and iterate autonomously, and let you know when it's "done".
I understand this probably does not work if you're on some subscription and not using the API (tokens burn fast), but this has been extremely productive for me.
Re: Backpressure is all you need
#10I’m willing to be wrong but this industry-wide emphasis on AI creative/coding workflows seems way over-engineered. Ime successful creative execution looks like micro-iterations where each output informs the next creative move. I can build something incredibly fast from essentially caveman grunt instructions through an LLM harness, iterating as I go. Optimizing for feeding a huge plan to an agent sounds to me like a n…
you do not need micro iterations. you can set macro goals and let the agent/LLM/model whatever you want to call it figure it out.
it works.