SynqoraX AEON
SynqoraX AEON
Menú

How the AI Actually Works

  • Machine Learning Model Layer Models are trained on historical and live crypto market data, then updated as conditions shift, so pattern recognition adapts across both spot and futures markets.
  • Signal Generation Process Pattern analysis is converted into concrete trade signals, each ranked by confidence, and passed directly to the execution engine without manual intervention from the trader.
  • Backtesting Before Deployment Every strategy is tested against historical market data before it enters the live library, so its behaviour under past conditions is understood before real capital is committed.
  • Server Side Execution Logic Orders are placed and managed on our servers rather than the trader's device, so a running strategy continues through a lost connection, a flat battery or a power cut.
  • Risk Limits Applied Before Trades Position size and exposure are checked against configurable limits before an order reaches the market, not after, keeping every trade inside the boundaries a trader has set.
  • Machine Learning Model Layer Models are trained on historical and live crypto market data, then updated as conditions shift, so pattern recognition adapts across both spot and futures markets.
  • Signal Generation Process Pattern analysis is converted into concrete trade signals, each ranked by confidence, and passed directly to the execution engine without manual intervention from the trader.
  • Backtesting Before Deployment Every strategy is tested against historical market data before it enters the live library, so its behaviour under past conditions is understood before real capital is committed.
  • Server Side Execution Logic Orders are placed and managed on our servers rather than the trader's device, so a running strategy continues through a lost connection, a flat battery or a power cut.
  • Risk Limits Applied Before Trades Position size and exposure are checked against configurable limits before an order reaches the market, not after, keeping every trade inside the boundaries a trader has set.
  • Machine Learning Model Layer Models are trained on historical and live crypto market data, then updated as conditions shift, so pattern recognition adapts across both spot and futures markets.
  • Signal Generation Process Pattern analysis is converted into concrete trade signals, each ranked by confidence, and passed directly to the execution engine without manual intervention from the trader.
  • Backtesting Before Deployment Every strategy is tested against historical market data before it enters the live library, so its behaviour under past conditions is understood before real capital is committed.
  • Server Side Execution Logic Orders are placed and managed on our servers rather than the trader's device, so a running strategy continues through a lost connection, a flat battery or a power cut.
  • Risk Limits Applied Before Trades Position size and exposure are checked against configurable limits before an order reaches the market, not after, keeping every trade inside the boundaries a trader has set.

How our AI works

Synqorax AEON applies machine learning models to crypto market data so that a trader does not have to watch charts around the clock. This page explains, in plain terms, how our AI trading works from the moment data arrives to the moment an order reaches the market, and what the system deliberately does not attempt to do. The goal is a clear picture of the mechanics, not a promise about outcomes.

What data goes in

The models behind Synqorax AEON are built on market data: price action, order book depth, trading volume and volatility across the spot and futures markets the workspace supports. This data is processed continuously so that the system is always working from a current view of conditions rather than a stale snapshot. Historical data is used during development and testing, so that a strategy’s logic can be examined against many different market regimes before it is made available inside the workspace.

What goes into the models is deliberately kept to information that describes the market itself, not speculation about news events or social sentiment that can be noisy and hard to weigh reliably. This keeps the inputs consistent and auditable internally, which matters when a strategy is later reviewed, adjusted or retired. The workspace does not ask a trader to supply their own data feeds; everything a deployed strategy needs is already built into the pipeline that feeds the models.

What the models are actually doing

At their core, the machine learning models used by Synqorax AEON are pattern-recognition systems. They are trained to identify recurring structures in price and volume behaviour, such as how a market tends to behave after certain shifts in volatility or order flow. This is signal generation in the practical sense: the model produces an assessment of current conditions relative to patterns it has seen before, not a prophecy about what will happen next.

Each strategy in the library is built around a distinct set of these patterns, which is why the strategies behave differently from one another in the same market conditions. Development work relies on backtesting against historical data to check that a strategy’s logic behaves as intended and to catch structural flaws early, but backtesting is a design tool, not a forecast of future results. We do not present backtest figures as a preview of what a live account should expect, because market conditions change and past patterns do not repeat exactly.

From signal to order

Once a model produces a signal, the execution logic decides whether and how that signal becomes an order. This includes sizing the position within the limits the trader has set, choosing between the long and short directions the strategy is built for, and timing the order so it interacts sensibly with current market conditions rather than firing indiscriminately. This layer is what allows a strategy to keep running unattended: because execution happens on our servers rather than on the trader’s device, a strategy and its logic continue operating through a dropped connection or a closed browser tab.

Execution logic also handles the administrative side of a trade: recording the fill, updating the position, and feeding that information back into the reporting a trader sees inside the workspace. Nothing about this stage requires the trader to be present, though alerts can be configured so they are notified when a fill or a limit event occurs.

The risk gate before execution

Before any signal is allowed to become a live order, it passes through a risk gate. This is a separate check that applies the limits a trader has configured for that strategy, including position size and exposure boundaries, and it runs before the order reaches the market rather than after. If a proposed order would breach a configured limit, it is not placed. This ordering matters: risk controls that only review a trade after execution can only report a problem, while a check applied beforehand can prevent it.

Risk limits in Synqorax AEON are configurable per strategy, which means a trader can run more than one strategy at different risk tolerances within the same account. The risk gate does not evaluate whether a signal looks promising; it only evaluates whether the resulting order fits within the boundaries that have been set, which keeps the two decisions, generation and control, separate from one another.

What this approach cannot do

Machine learning models describe patterns in past and current data. They do not predict the future, and no configuration of Synqorax AEON changes that basic fact. A model can identify that a certain condition has historically preceded certain behaviour, but markets shift, liquidity changes and new conditions arise that no historical pattern captured. Treating a signal as a certainty rather than an informed estimate is a misunderstanding of what the system is doing.

Trading digital assets carries substantial risk, including the total loss of capital, and nothing on this website is investment advice. Past performance and any illustrative figures do not guarantee future results, and a well-built risk gate reduces certain kinds of exposure without removing the underlying uncertainty of the market itself. Anyone using Synqorax AEON should treat the models as a tool for structuring decisions, not as a substitute for accepting that outcomes remain uncertain.