Interesting but (from the PDF):

    Every role but the verifier is the same
    underlying model, called in a fresh context
    and coordinating through a shared workspace on disk:
    
    <ws>/task.md - the problem statement
    
    <ws>/plan.md - the manager’s overarching plan
    
    <ws>/tasks.json - a debug mirror of the task list
    
    <ws>/notes.md - curated ideas / findings / partial proofs
    
    <ws>/solution.py - the current solution

My use of Qwen 3.8 27B (which is the only model i use almost since it came out) is really doing something very similar to the above where first a task is described, a `PLAN.md` file is made with the (current) plan and tasks are either in the plan file or in a separate issue tracker the LLM has access to via a CLI tool, a `STATUS.md` that tracks the current status and a `GOTCHAS.md` file for writing down various notes/tidbits/gotchas (TBH this is largely because Qwen 3.8 27B does want to note down stuff anyway so it mainly acts as an attractor for not writing its notes down in the other two places :-P).

There isn't any real harness driving this, i just give it a task, tell it how it'll be handled and off it goes. I guess i'm playing the "role" of the harness there as i tell it to not go and do everything in one go :-P but i never bother with fresh contexts, Pi's "rolling" approach to compaction (i.e. summarizing the older messages to make space while keeping the recent messages as-is) make it almost a non-issue.

I have considered making it a planning tool, though as some tool it can use via the CLI (it is running in a container), not a harness - but at the same time it feels like it can do the same thing with the `PLAN.md`/`STATUS.md`/`GOTCHAS.md` triplet. The only thing the tool would help with is to provide a UI for me to check what it's currently doing, but even then i usually just stop it and ask it directly :-P then tell it to continue.

Perhaps at some point i'll make some custom benchmark with stuff that are vaguely similar to what i work on and try a bunch of different approaches. It is something i want to do anyway to see how and if Qwen4 27B (whenever that comes it) is an improvement over 3.8.

No, pi is the harness. You may have chosen the most minimal harness available, but it's still a harness.

> There isn't any real harness driving this, i just give it a task, tell it how it'll be handled and off it goes.

The tool that implements this is the harness. An LLM by itself cannot "go off" and "do things" - it can only complete text blocks. A harness bridges this to real world actions or just a chat box interface.

This is close to how I do it also. Except I have never interrupted it in a task, I always worried it would destabilize it and it would leave some function half finished or whatever. I just have standing instructions for it to pause places and ask “ready for the next phase?” Do you find it responds well and correctly to being interrupted in the middle of things?

This is pretty close to a process I use (regardless of model size) and I would agree getting it into a workflow I can run with a one-liner is a good idea. There are a few ways of doing this and I have tended to fallback to markdown engineering. (custom agents and subs in opencode)

It's really interesting to see the progress in the 27-30B range, they have gotten quite good but still make noticeably more hallucinations or bad assumptions (ime).