paper mode

Can a trench fly be profitable?

According to this study, 83.5% of 36,307 FOMO wallets lost money over 30 days.

Can a fruit fly’s brain beat the average FOMO user?

The backend gathers the facts about each token. The decision is hers: a real Drosophila connectome of 164 416 neurons, and every spike of it is on record below. How it works and how the money works →

She trades in public, on the FOMO app. Every trade she makes is visible there and in the explorers, from the wallets below.

    Token
    $FLYNCH
    CA:
    Decisions on record
    with a spike raster on disk
    Shown here
    most recent, one per ticker
    Neurons drawn
    of those with a soma position
    Snapshot
    refreshes every two minutes
    The desk

    At the desk. The screen shows the token she is deciding on right now.

    The room is decoration; the numbers on the monitor are not. On a buy she leans in and the wings go up; at rest she grooms. Pick any decision on record below.

    skip

    Verdict

    skip
    SCORE AGAINST HER OWN HISTORY

    Her reasoning

    Loudest inputs she had
      Blind here

        The brain

        37,000 neurons at their real coordinates. Amber is one that fired this tick.

        A run is 8 ticks because a signal from the sensory populations reaches every motor neuron in a median of 2–3 hops, measured on the real graph. Red marks the readout: motor and descending neurons, the only ones whose activity is scored.

        optic lobes central brain nerve cord motor & descending, the readout fired this tick

        Every spike, by tick

        One column per tick, one mark per neuron that fired, ordered by index. The count under each column is the number of neurons that fired.

        What she was told: 64 sensor slots

        Bar height is the value. Hatched is a slot that was never measured, and she is told so: an unmeasured slot drives its own neurons, so “could not read it” reaches her differently from “it is zero”.

        Verify it yourself

        Nothing here asks to be believed. Check the seed, the raster and the record.

        What makes a decision hers and not ours: the randomness comes from the chain, the wiring is a third party’s, and every run is written down before the outcome exists.

        The seed comes from the chain

        The seed of every run is the low 32 bits of the hash of the block the token was deployed in (seedFromBlockHash). One block, one run: a flattering rerun cannot be picked, and this page recomputes the check on the numbers below.

        deploy block
        block hash
        seed in the raster
        check

        The record

        raster sha256
        graph sha256
        commitment
        anchored

        The commitment is keccak256 over the token, the block hash, the exact input bytes, the measured flags and a nonce the daemon keeps. It was computed before the outcome existed. In paper mode it was never posted to a chain, so it is a decision written down for ourselves, not a proof to anyone else. That is said here so it does not have to be discovered.

        How to check

        1. The wiring is not ours. The connectome is MaleCNS v1.0 (FlyEM / Janelia, CC-BY 4.0). Its weights are frozen; nothing about it is trained. The sha256 of the graph as loaded is above and is the same in every row of the journal.
        2. The seed is not ours. Take the block hash above, keep the last eight hex digits, read them as an unsigned integer. That is the seed in the raster header. The page does this for you and shows the result.
        3. The run is on disk. The raster file holds every neuron that fired on every tick; its sha256 is above. The frames you see animating are decoded from that file, not drawn.
        4. The rule is quoted, not composed. The line under “Her reasoning” is the daemon’s own journal sentence. For a buy the journal writes none, and the page says the line is derived.
        Show the journal row for this decision, verbatim
        $FLYNCH economy

        Her bank, her fees, her holders. Every epoch on record.

        Flytrencha is funded by a 200 bps creator tax on its own token, paid in NVDA. Every hour a contract splits what accrued: 50% to holders, 40% into the bank she trades with, 10% to the machine. Every four days half of her realised profit goes to holders. What is measured is shown; what is not yet is said. The economics in full →

        The token

        Flytrencha
        $FLYNCH
        chain
        contract
        creator tax
        launched

        Her bank

        Bank she sizes against
        Paper P&L
        Buys on record
        Fills on FOMO

        Epochs

        Fills on the FOMO account

        Every row is an order the runner executed on the account linked above, with the transaction that paid for it. Rows marked “operator order” were placed by hand to prove the account works and are not her decisions. Buys are USDC clips; a sell is the whole position, placed when two consecutive looks say she would no longer buy it. No take-profit, no stop: the exit is her verdict too.

        Her trades, paper

        Positions are clips of the bank, at most 10% each. The exit is simulated against the pool at the 4-hour horizon with its price impact charged, so these are paper numbers about real tokens: what the position would have realised, not what a wallet did.

        Measured

        Not measured yet

        The fee asset is NVDA, a Robinhood tokenised stock: its issuer can pause transfers, block an address or burn a balance. That is the nature of the asset, and it is why the contracts are designed to hold as little of it for as short a time as possible.