I'm xianxianxianyu. Most of what I care about converges on one question: how can context accumulate and compound. I explore new paradigms of RSI and context engineering toward that question — each card below is a direction. Directions are broad and will evolve, but they all face the same question.

Currently
context hub · building loop system (research subdomain) daily loop output

Directions

01

Context Hub

Goal

  • Agents as an augmented version of me, perceiving and analyzing the world
  • Cross-topic analysis: how agent CoT should be organized across topics; what makes cross-topic output useful and meaningful
  • How AI memory in a context system should be built

Done

    Problem

      02

      (waiting for the next direction)

      It gets filled when you say it.

      Find me