Strategy brain, 2D fallback: up to six lobes (dormant ones outlined) with strategy nodes and composition nerves
What is CORTYN?

It starts simple and grows by selection. It began as a small 3-region brain; many AI models now compete on paper in a tournament, the weak are culled every generation, and only the best few get real money. As they prove themselves, new brain regions (provenance, flow, curve) unlock one by one, shown as outlined embryos until they are born.

This is early. Real money is tiny, most stages of the roadmap are not reached yet, and real and simulated results are never mixed.

Neural circuits (shadow). A circuits view shows eight named circuits and the messages between them, from recorded state only. It makes recommendations, never trades, and says so when it is not enabled.

Risk: you can lose everything you put into anything crypto. This is research, not financial advice. The CORTYN token identity and public research sources are linked in the footer. Naming the token is not an endorsement or a promise of value.

0 signals · 60 s · this browser
    LIVE
    Legend
    • active strategy · size = √ paper trades
    • challenger
    • eliminated (ghost)
    • composition tract (paper) · brighter = open position · thicker = |paper PnL|
    • inside a region: each competing strategy gets its own hue; its tracts share it
    • parent → child lineage
    • pulse: paper fill
    • pulse: decision
    • pulse: observation (lime live · pale recorded · grey synthetic)

    Drag to orbit · scroll to zoom · hover a region, click to enter · Esc goes back

    Loading the specimen.

    telemetry loading state.

    populationsaccessible list · same data as the brain

    Loading.

    paper taperecent fills

    Loading.

    observed tokenslaunchpad feed

    Loading.

    generationstournament log

    Loading.

    paper equityPAPER

    Loading.

    read this first

    Paper first, tiny real trades second. Almost everything you see is simulated against quoted prices. A small hot wallet can also place real trades on Pons (Robinhood Chain), only inside hard limits: about 0.0004 ETH per trade, 3 open positions, daily-loss and drawdown halts, and a kill switch. Real trading is currently halted by the drawdown limit and net negative so far. Real transactions are listed with explorer links in the Live Log.

    Rule strategies and AI models. The core strategies are plain, versioned code with seeded mutation. On top of them, AI models (called through Surplus Intelligence) paper-trade in a tournament, and their calls are logged and costed on the economy view. AI proposals never bypass the deterministic risk gate.

    Every number comes from the local API. The only figures counted in this browser are the signal rate and the equity trace since you opened the page, and they are labelled. Pulses on the brain appear only when an event is delivered; when nothing arrives, the brain only breathes and says IDLE.

    LIVE
    observed on Robinhood Chain (Pons) just now
    RECORDED
    genuine network logs, replayed. Not live.
    PAPER
    simulated fill. No real order exists.
    SYNTH TEST
    generated test data. Not market data.
    POLICY
    output of a deterministic strategy
    HISTORICAL
    structure of an earlier generation, from the tournament log