Eigenstate

Jason Shen / Trading infrastructure · Human-agent collaboration

Inspectable systems for trading and human-agent collaboration under pressure.

I build systems where internal state is not allowed to pretend it is truth.

Trading infrastructure taught me one hard rule: intent, state, risk, and confirmed truth must stay separate enough to be reviewed. Battuta is where that discipline proves it travels — from financial systems to musical decisions.

On the record

01
artifact you can open
Trading State Replay — execution-state drift, replayable
01
essay published
Ledger opened 2026-05-24
02
systems on the record
Trading infrastructure · Battuta
SG
Singapore, GMT+8
Quant developer — financial systems engineering and AI agents

Systems

Two systems, one rule: state must stay reviewable.

Writing and lab

I write the reasoning down before the playbook exists.

Three surfaces, by how settled the thinking is: Ansatz for field notes, Eigenvalues for essays that have to hold up, Impuls for experiments you can run.

The first essay is live: Maxwell's demon as a sharper metaphor for market efficiency — information cost, and trading edge as state classification.

Collaboration entry

Three kinds of problem where a conversation is worth it.

Best fit:

  • trading, execution, or risk systems with unclear state boundaries
  • human-agent collaboration where the boundary between tool and judgment matters
  • technical partner conversations around infra-heavy products
Start a collaboration note

Architecture lens

Every system here runs the same loop.

source → core → validation → feedback