The desk doesn't stop
when you close the tab.
Autopilot is the research loop running on a schedule: analysts file, their boss cuts, the executive rules, and every round is written down whether you were watching or not. It trades a simulated book. It has no path to a brokerage, so it cannot touch your money, not by permission, by construction.
Simulated book · no broker connection · not advice.
Watch it trade, in real time, with fake money.
Every fill below is Flux acting on its own inside a $10,000 paper account. No real orders, no real dollars, just a transparent record of what the AI decided and why, so you can judge it before it ever touches your book.
Mirrors the fund's current stance into your own simulated $10,000 paper account, equal-weight, in virtual money. Simulated, not advice.
| Time | Ticker | Side | Qty | Price | Reason |
|---|
The Flux Fund is a simulated $10,000 paper account using virtual money. Every figure here is hypothetical and does not represent real trading or real performance. Simulated results have inherent limitations and are not indicative of future results. Flux is not a broker-dealer and none of this is investment advice.
Pick a rule. Watch it play out, on synthetic prices.
This is the same idea behind Autopilot, made transparent: choose a simple, readable rule, see exactly which names it holds right now and why, then run a seeded 30-day paper simulation on a synthetic price path. No real money, no live orders, just the concept, in the open.
Simulated backtest on synthetic price paths, for illustration only. This is not a real account, not a prediction, and not investment advice. Results are generated from a seeded random walk, past or simulated performance is not indicative of future results. In live use, Flux Autopilot only ever acts on rules you set and approve, and you keep the stop button.
—
| Ticker | Name | Why it's picked | Wt. |
|---|
| Side | Ticker | Qty | Price | Value |
|---|
The simulator starts a hypothetical $100,000 paper account and marks it to a seeded, synthetic random-walk price path over ~30 steps. It is a demonstration of a transparent rule, not a backtest against real market history and not a forecast. Simulated results have inherent limitations and do not represent real trading or real performance. Flux never promises returns and this is not investment advice.
Every call walks the same five steps.
Nothing skips a desk and nothing reaches you until the last one has signed it, including the calls that get thrown out. Each step is written down.
Pick the names
The round opens on a slice of the desk's universe, with the tape's own rating and anything that crossed the wire on those names attached.
Three analysts file
Fundamentals, catalyst and tape. Each works one lens, files a direction, a target and a date it can be scored on, and never sees the other two.
Their boss cuts
The traders read every filing against that analyst's own hit rate and pass on only what they would put size behind. Most filings die here.
The executive rules
Approve, cut the size, or refuse, with the reason and which trader it sided with written down. It refuses far more than it approves.
The round is filed
Every filing, ruling and reason is stored. When a prediction reaches its date it is graded against the tape, and the grade goes on the scoreboard.
Model output can be wrong, and a filing is a hypothesis rather than a recommendation. Everything described here happens inside a simulated account with virtual money.
It never trades quietly.
Every market hour and every close, Flux writes up what it did and why. The transparency is the point; you should be able to read a full day of decisions before you trust it with a cent.
Hourly pulse
What it did this hour, the reason behind each fill, and current exposure across the paper book — filed on the top of every market hour.
Daily wrap
Every trade of the session, the day's simulated P/L, the best and worst positions, and the plan the AI has drawn up for tomorrow.
Weekly review
The long look: drawdown, correlation, rule adherence, and whether the edges the fund is trading still show up in recent data.
A worked example of the wrap's shape. It carries no figures, because this account has not traded. Yours is written off your own paper book at the close of each session.
The close. Where the paper account finished, the day's simulated move, and how many fills it took to get there.
Best and worst. The position that carried the day and the one that cost it, with the reason each went the way it did.
Exposure. How much of the book was at risk through the session, and every trade the Risk engine refused to let on top of it.
Tomorrow. What is being held, what is being tightened, and the one scheduled event that could change the answer.
What is actually running.
The trade log is the last step. Behind it, six specialised models each answer one narrow question, and the fund only acts when enough of them agree.
These are the models in the loop, named for what runs. Output is probabilistic and can be wrong; it is not a prediction you should act on, and it drives a simulated account only.
Guardrails you set, and can pull anytime.
The paper fund runs on its own. Your real account never does unless you say so, on rules you write, with a stop button that's always yours.
Never without approval
Flux never touches your real account until you explicitly opt in. Until then it only trades its own simulated paper money.
You set the rules
Max position size, daily loss limit, allowed tickers, sectors to avoid; you write the mandate and Autopilot trades inside it.
Pause or stop anytime
One tap freezes new trades or flattens the book. There is no lock-in and no waiting, control is always in your hands.
Every action is logged
Each decision, fill and reason is written down and reported hourly and daily. Nothing happens off the record.
Hard risk gates
The Risk engine blocks oversized, over-correlated or over-exposed trades before they're placed — on paper and, if enabled, on your account.
Kill-switch on drawdown
Hit your daily loss limit and Autopilot halts itself automatically, then tells you why. The floor is a rule, not a hope.
Counts above are read from the desk's own record. There is deliberately no uptime, win rate or average hold here: the desk trades a simulated book, so any such figure would be hypothetical. Figures were previously illustrative and describe activity inside a simulated $10,000 paper account. They are hypothetical, do not represent real trading or performance, and do not predict future outcomes. Flux never promises returns.
The fair questions.
Is this real money?
No. The Flux Fund is a simulated $10,000 paper account using virtual money. It exists as a transparency tool, a live demonstration of how the AI thinks and trades, not as a real portfolio. Every number you see is hypothetical.
Will it trade MY account?
Only if you opt in and set the rules. By default Autopilot trades nothing but its own paper money. If you connect a brokerage, you define the mandate, max size, loss limits, allowed tickers, and you can pause or stop it at any moment.
Can I see every trade?
Yes. The live log shows each fill as it happens, and the hourly and daily reports write up what the fund did and why. The whole point of the paper account is that nothing is hidden.
Is this financial advice?
No. Flux is research and automation software, not a broker-dealer and not an investment adviser. Nothing it produces is personalised advice or a recommendation, and every decision on your real account is yours.
What strategies does it use?
A multi-engine mix: momentum and breakout scanning, options flow, catalyst-driven setups and mean-reversion; each signal cross-checked and then risk-gated for size, stop, correlation and exposure before it's allowed to trade.
Can it lose money?
Yes. The simulated results include losing trades and losing days, because that's honest. Trading involves risk of loss, including loss of principal, and no strategy, human or AI, removes it.
Watch Flux trade. Then decide.
Follow the paper fund for a week, read the reports, and only then connect a brokerage, on your rules, with your stop button.