Hawaii's artifacts have been displaced far and wide, The British museum alone has a stunning collection that usually sits in a basement, e.g. a small sample of their collection documented here:

https://www.artoftheancestors.com/blog/hawaii-british-museum

[0] Building a hubless bucky ball dome. It's a bit deceptively-simple, because the agent only has to call the procedure with the right values. That's an intended flywheel: agents and users can save procedures to a community library, and future agents can discover and invoke the pre-existing procedures, instead of re-deriving them. They can copy and edit the procedure as needed, so it basically becomes a RAG library for carpentry procedures written in the project's DSL.

[1] A sawhorse, re-implemented by Opus 4.8 from plans published by Home Depot. Now saved as a procedure so you can drop a sawhorse anywhere you need one.

[2] A small shed with stick frame walls and a lean-to roof.

[3] The Picture frame prompt mentioned in my first comment.

Quick note: I recently broke the smart de-duplication for step-by-step playback, so it will show every single cut, one by one. Hit "stop" and then click the final frame in the timeline to see just the finished state.

[0] https://sawdust.diy/share/b0de719c-0e9f-4f4f-9282-085c521163... (bucky ball dome)

[1] https://sawdust.diy/share/45557307-0f78-4b5b-bf0e-eae77a9853... (sawhorse)

[2] https://sawdust.diy/share/cbb591fc-5511-40d9-8ccd-0d9f8c10be... (shed)

[3] https://sawdust.diy/share/ba109216-8849-49e3-ae9c-5a15982d24... (polygonal frames)

There is a chat window and human UI similar to SketchUp. You can task the agent with using the tools and assembling the project, or do it yourself. It outputs real cad files, plans, and a build guide video.

Qwen 3.6 has been great because it's multimodal and good at tool calling. As it builds it can get screenshots and all that. The basic output is an operation list with real measurements and bevel/angle settings and the like.

I use Qwen3.6-A3B locally on my Strix halo for testing simple prompts like "make the largest polygonal picture frames you can, each from a single 8 foot 2x4. Start with 3 sided and go up from there, lay them all out in a 9 by 9 grid."

It's a cheap (off-grid solar power, so ~free) way to sharpen the harness: if less-than-SOTA models can succeed then the tool surface is clear. Bumping up to 3.6 Pro has been a nice middle step closer to the frontier models without spending an arm and a leg on credits.

I ended up thinking it might be easier to generate rigged models, animate them, and capture from an iso perspective, then do some kind of pixel art style transfer on the masked sprite sheet. Eventually I realized my kid didn't really care too much about the visuals so I didn't get too far with it.

Theory 1: Some increasingly-large split of inference compute is moving over to serving the new model for internal users (or partners that are trialing the next models). This results in less compute but the same increasing demand for the previous model. Providers may respond by using quantizations or distillations, compressing k/v store, tweaking parameters, and/or changing system prompts to try to use fewer tokens.

Theory 2: Internal evals are obviously done using full strength models with internally-optimized system prompts. When models are shipped into production the system prompt will inherently need changes. Each time a problematic issue rises to the attention of the team, there is a solid chance it results in a new sentence or two added to the system prompt. These grow over time as bad shit happens with the model in the real world. But it doesn't even need to be a harmful case or bad bugged behavior of the model, even newer models with enhanced capabilities (e.g. mythos) may get protected against in prompts used in agent harnesses (CC) or as system prompts, resulting in a more and more complex system prompt. This has something like "cognitive burden" for the model, which diverges further and further from the eval.

alias yt-pl='yt-dlp -o "%(channel)s/%(playlist_title)s/%(title)s.%(ext)s" -a playlists.txt'

I am a software engineer, I also runs a small family farm. I have 3d printers and laser cutters and lots of aluminum extrusion and raspberry pis... but I keep those things indoors, away from the dirt, sun, and rain. I can't imagine a real farmer using a contraption like this. Tools have to be reliable to last. I have to replace my solid steel shovels every few years because they wear out, how is this supposed to work?

How many of these passing tests are still actually testing anything? Do they test the tests?