LFS prepared me for Gentoo :)

And then another sub agent that argues for the whole system to be re-written in another language

If that's your goal, then yes. Invoking sub agents (with a fresh context) corrects most of these problems. Ask your harness to create a commit gate.

But why stop at rewriting in another language. Get another sub agent to invent a new language, create a database, query language and maybe another few DSLs. Then you've got an ecosystem!

You can now re-position your initial solution and sell the client access to some agents that will implement & configure the ecosystem to suit their initial needs!

And don't forget the agents that you'll need to train the customer to use the whole thing!

I know you're trying to be funny - but I'm offering a real fix for his problem.

If you don't want a million agents arguing about things, you simply don't ask for that. One agent is sufficient to solve most issues.

Sorry, I wasn't implying your advice doesn't carry weight. Was more just thinking about the things that (used to) happen when you introduce more parties to process of creating software.

I wonder if I would need a non-openai agent to enforce it.. I have tried so far with skills and agents.md and code stills end up over engineered to the moon.

Will ask OpenAI to write me that agent! Hope the agent is not over engineered or else unsure how to solve the bootstrap puzzle :D

There's a misconception that a mistake one instance of a model makes will be repeated by other instances. In my experience most mistakes are just sampling noise, and having the same model (with empty context) check the work usually fixes it.

It's super easy - just ask your agent to set up a code review commit gate.

The voices in my head argue about the direction of the project enough already

Would you be able to taper by stretching out dosages, take a pill every 25 hours instead of 24?

Just don't put it in charge of a company making paper clips

Would building an LLM from scratch imply writing code by hand?

> circular economy

I get what you mean, but in general circular economy means something different, positive in my opinion. Here is what wikipedia has to say:

> a model of resource production and consumption that involves sharing, leasing, reusing, repairing, refurbishing, and recycling materials and products, to extend product life cycle for as long as possible

In the EU it is a hot topic for a lot of non-profit and social innovation work

I think the term GP was looking for was "circular financing", which has far fewer positive connotations.

Right. Almost every economy is circular in some fashion. Every producer is also a consumer. Sometimes the circle is quite big, but many of the dollars (or other currency) that you spend, come back to you in the money you earn.

Many times I wonder if it's possible to collapse abstractions? The idea of making the whole stack simpler. I'm not overly optimistic of LLMs achieving this, but it they do, I will be paying more attention.

This immediately made me think of nand2tetris.org, but as a game. Is that what NANDGame is?

Yes, pretty much, but stopping at the cpu level (about the first half of the course).

I'm curious about ways to entice agents to perform weird and wacky things when working on a codebase.

Isn't the problem with formal methods that it isn't clear whether or not most of the useful code we use are actually formally verifiable?

The article mentions NP-complete, but is it actually a solvable problem in general?

> For extremely restricted cases, like propositional logic or HM type-checking, it’s “only” NP-complete.

No, lots of problems can be expressed in a way that can be verified. But complete verification of an existing implementation is essentially impossible.

That doesn't mean that formal techniques are not useful, far from it. For example, AWS uses a formally specified model to verify if an implementation is correct by looking at the telemetry. See e.g.

https://p-org.github.io/P/advanced/pobserve/pobserve/

This isn't something you could meaningfully do with standard testing techniques, and it very compositional, you can do it piece by piece.