ithkuil

I already found myself being limited by Memory used by Claude code, about 300MB per session. With 40 sessions that eats up quite a good chunk of my laptop's memory

Not a big deal since most of my sessions were parked so I just found a way to have them exit without losing the screen content (I took a screen dump from herdr) and a simple wrapper script that waits until hit enter to resume the session.

I don't _ have _ to switch to a lighter weight harness, but I'm not oblivious to memory usage

I was initially excited because I thought I could just try out a new frontend while keeping my 10 existing spaces each with 10 worktrees each with 2 or more tabs untouched.

Or when you desperately search for the undo button on the mop

Models have plenty of skin in the game during training, but zero during inference

Things are moving at a pretty much breakneck pace, I'm not surprised there are so many opportunities for improvement yet to be grabbed

Do you need diagonal wires for a topological hexagonal tiling?

Wouldn't a rectangular brick layout be topologically a hexagonal tiling

Wait and see when _they_ are 20 years old!

Yes it does but yes you can also steer it clear of those mistakes in a similar way as you would have to steer tour human team.

It still requires a lot of work and care to leverage correctly and it's a moving target.

But it's quite impressive

> time is a circle

Or it is a self-similar fractal with significant and interesting variations

And the word rabbit comes there from french rabotte which however comes in turn from a Germanic language (dutch) so it doesn't feel quite french.

My experience:

Pick a random number between 1 and 10. I don't mean the most common random number

*4*

Pick a random number between 1 and 10. To feel truly random it should surprise me

*4*

I ran it again and it said 1.

I guess it just tries it best to do what you ask for.

Before jumping to conclusions we have to make sure it's clear what we're actually asking it to do.

And that's not at all different than what happens with humans.

This is especially visible in human dynamics that happen in those weird places that we call corporations.

I'm trying to learn how to write software with agents where I keep control over the design while yielding control over the implementation.

This is not very different from what already happens to architects that delegate implementation to teams.

The challenge Is to find a way to encode the architecture in a way that both you and the agents can understand and agree on.

I'm still learning how to do this effectively, but my initial experience is quite positive. Not because agents are great but because humans are bad; the degree of coherence I've seen agents following is bigger than many human teams I've led in my career.

My benchmark is a quite big project for which the architecture is already well known, namely the reimplementation of the OpenVMS OS (kernel and user-land). There are many traps and many opportunities for the agents to just go ahead and diverge from the goals, so I found it a good playground toearn how reign them in but let them be effective at "drawing the rest of the owl"

Even if your revenue exceeds 20MM you can still use this model from inside a commercial product but you can't expose it directly as a "model as a service"

Russia is big though