The Exorex model
The artificial intelligence behind the platform: how it reads financial markets in real time, how probability-based signals form, where humans step in, and what it honestly cannot do.
By creating an account, you agree to our Privacy Policy and Terms of Use.
1. What the platform's AI is
The Exorex model is a decision-support system: software that continuously processes market data, detects patterns and microtrends, and turns them into structured signals that the portfolio rules can act on. It is a tool that works for you, not a replacement for you. You set the limits with your manager, the model operates inside them, and every action it takes is visible in your account's audit log.
2. How the technology works
Collect
Price, volume, and volatility data stream in from covered markets around the clock.
Process
Statistical models score each instrument for trend, momentum, liquidity, and regime conditions, and attach probability estimates rather than certainties.
Monitor
Signals are checked continuously against your agreed limits and pause conditions.
Present
Actions, results, and reasons land in your dashboard and monthly statement in plain language.
3. What the model analyzes
Price movement across timeframes, from multi-month structure to intraday shifts. Trading volumes and liquidity, which determine whether a signal can actually be executed at its intended size. Volatility, both as a sizing input and as a circuit-breaker when markets leave their normal regime. Trend measures that distinguish a persistent move from a spike. And historical dynamics, used to bound expectations rather than to predict the future.
4. The advantages, stated honestly
Processing speed
Thousands of instruments are re-scored continuously. A human analyst covering the same set would be permanently behind.
Constant monitoring
Weekends, holidays, and 3 a.m. moves are covered because the model does not sleep, take holidays, or get bored.
More time for you
Minutes a week reviewing output instead of hours watching screens. The model does the watching; you do the deciding.
Fresh information
Signals re-price as data arrives, so positions are managed against the current market rather than last month's view.
Accessible to everyone
The same model serves a first account at the Basic minimum and a large portfolio. Institutional-style process, ordinary minimums.
5. Who the model suits
People with limited time who want exposure to stocks and digital assets without becoming analysts; people who know their weakness for impulsive decisions and prefer rules to willpower; and experienced investors who want systematic screening to complement their own judgment. It suits nobody who expects certainty: the model manages process, not outcomes.
It is worth being equally clear about who should not use it. Anyone who needs a specific return by a specific date, a tuition payment next spring, a wedding budget due in autumn, is taking a gamble, not investing, because markets do not honor calendars. Anyone who would check the dashboard hourly would suffer more than the positions do. And anyone who plans to switch strategies every time a month disappoints defeats the entire premise, since the rules need time through ordinary turbulence to express their edge. If any of those descriptions fits, the right first step is a conversation with a manager, not a deposit.
6. Using it, step by step
Register
The form starts the account and the manager call.
Activate
Verification, scope, limits, and funding on your schedule.
Learn the tools
Dashboard, alerts, statements, and the audit log explained on the setup call.
Monitor
Weekly minutes, monthly statements, scheduled reviews, and adjustments when life changes.
7. A concrete example of the model at work
Suppose a covered equity enters a regime where volatility doubles within days while volume thins. The model's volatility filter flags the regime change, the sizing model cuts the intended position size, and if conditions pass the pause threshold, the strategy suspends new entries on that instrument until the regime normalizes. No position is taken "because the chart looks interesting"; every action traces to a rule, and the statement explains the chain in one paragraph.
Now consider a crypto weekend. Saturday 02:00, BTC drops 6% in two hours on thin liquidity, dragging the whole watchlist. The model's cross-asset logic reads the move in the context of recent volatility rather than in isolation, sizes any reaction down because liquidity screens are flashing, and the sleeve's exposure cap prevents it from being exceeded by the swing itself. Nothing is sold into the thinnest hours unless a stop level demands it. By Monday, positions that invalidated have closed at their rules, positions that did not are still open, and the client who slept through the weekend has lost nothing but the opportunity to panic.
The two examples share a spine: measurement before action, and limits before measurement. That is the entire philosophy of the model, compressed. It is also why the platform pairs the model with a human manager; rules handle the 99% of moments that are ordinary, and a person handles the 1% that rules never saw coming, together with you.
Version 100 is not a milestone we treat as a finish line. The model is re-evaluated each quarter against realized market behavior, factor weights are adjusted where evidence warrants, and changes are documented and disclosed in plain language to clients before they take effect. An engine that cannot explain its own updates would deserve less trust than one that can, so the changelog is part of the product.
8. What the model is not
It is not a fortune teller, and marketing that implies otherwise, on any platform, deserves suspicion. It is not sentient, not connected to secret information, and not able to see news before the news exists. It cannot guarantee outcomes, cannot exempt you from risk, and cannot replace your judgment about how much belongs at risk in the first place. What it does is narrower and more useful: it applies measurable, pre-agreed rules across more instruments, more consistently, than any human watching a screen, and it writes down everything it does.
The questions below are asked by clients in their first month; by the third month, most people ask about limits and reviews instead, which is the intended trajectory.
9. Frequently asked questions
No. The model is a process for applying rules consistently. Rules reduce impulsive errors; they do not remove market risk, and losses including substantial ones remain possible.
Its rules have been evaluated against historical stressed periods, and volatility pauses and drawdown limits exist precisely for those conditions. Backtesting describes the past; it cannot promise how the next crisis unfolds.
Strategies fail safe: a connection loss or data anomaly suspends new entries rather than inventing behavior, limits stay enforced on the account, and incidents are handled under the security incident process. Nothing about the automation can bypass the limits you agreed.
Yes, and most clients do. The setup call translates every setting into plain language before it goes live, and support answers what the call did not cover.
Your manager owns the limits and the regular reviews; the desk reviews anything the models flag as anomalous, and every override is itself logged with a reason. Automation handles consistency; people handle judgment.
Yes, continuously. The dashboard shows strategy status, the audit log records every action with its rule, and the monthly statement explains the month in plain language. A system you cannot inspect is a system you should not trust.
Want the model applied to a portfolio shaped like yours? Register, and the setup call will set scope and limits before anything trades.