#thecompaniesapi signals feature in production; went from fully killing the CPU of multiple machines to <1% CPU usage; make it work then make it fast

finish aligning & processing great sphinx data w/headless metashape on a L4 after ~6 days #mused

try 3 different instances to improve musicgen generation time #ai

#thecompaniesapi working on llama3.1 with large context window, trying to improve some parts of our AI flow before merging it; now properly respects industries & naics constraints for current sample

try older model version and continue experimenting with optimizations #ai

fine tuned data cleaning to handle the limits on Claude 3.5 Sonnet Context Window.. now continues processing where it left off and we've added the ability to fact check and complete, incomplete datasets. I really want to add a diff view instead of just clobbering existing data, but we're getting into scope creep. Can come after launch. #datamachine

continue working on new idea #ai

realizes I need to scale down version 2.0 massively, so now making a clone of v1.0 with under the hood improvements to start with.

keep working on #oxbowui

keep working on #birkenwaldart

finetune import && make more stable

plan speeding up the CPU, increasing RAM and installing ISDN (double the modem speed) with @bai0 on pieter.com #pieter so we can get fast internet and fix slow Netscape and crashing Eudora email and make it more usable

keep working on prototyping #brewfolk

speed up #photoai to below 1000ms again

Work on 3rd prototype UI #jovial

Work on 3rd prototype UI #jovial

finished optimization experiement #blankpage

work on hpn #freelance

work on improving LP #shipr

continued work on the new prototype #blinkloader