Ang Li

@angli_ai

CEO @SimularAI | Creating autonomous computers @sai_borg | Former @GoogleDeepMind Research Scientist | The future of digital experiences

A post says the meta-agent reads task trajectories without success labels, then builds interfaces while the model stays fixed.

9 Sources, 5h ago, updated 5h ago

An October 15 post says @Richard_simular was presenting on the topic at BayLearn that day.

2 Sources, 6d ago, first seen 6d ago

Sai's team says OSWorld verified its run and cites a $14.34 cost per task, versus $24.11 for Anthropic's Claude Computer Use with Opus 5.

2 Sources, 16d ago, first seen 16d ago

Its introducer says stable knowledge becomes executable code, while perception and uncertain decisions remain neural.

5 Sources, 19d ago, updated 19d ago

Featured in