nikhizzle

Claude code

Kind of cool how fable 5.1 just worked around these by reassigning subagents to non-failing models with more supervision and now with a hint to Sol and Flash 3.8.

IMHO from guy working with heavily quantized models and trying to get results. I do think quantization at less than 4-bit requires more advanced mathematics than is usually used. See Google Scholar Babak Hassibi (founder of Prism ML). Generally am amazed about the resiliency of these new data structures to approximation.

Got to my first job out of college and they gave me core dumps and had me debug the kernel for a year. Not my kind of fun, but definitely got me skilled in the art of low level.

I believe I potentially wrote that code in 2006 when working at Apple on HFS. What a remarkable world!

Ex-Apple kernel engineer here, Apple will deal with the memory shortage by making software more efficient in ram usage. Apple will just make every aspect of the system more and more memory efficient. They've done it before over and over and can do it again.

Almost trivial using open source tools, the question is how it performs without calibration/fine tuning.

Will think about it, trying to keep it running snappy with minimal cost since no clear path to monetization