We apply AI and machine learning to read markets and forecast price moves — first for our own trading, and in time as a product for others.
Our models exist for one purpose — to make better decisions on the company's own capital.
Continuous ingestion of prices, volumes, volatility and correlations across markets.
Machine-learning models separate structure from noise and surface what matters.
Models estimate the probability and direction of price moves across time horizons.
Each forecast is checked against risk limits before it becomes a position.
There is no single magic model. It's an ensemble — statistical, neural and probabilistic — each good at its part of the problem. The system retrains continuously on new data and adapts as the market regime shifts.
The AI we build to trade our own capital is, in time, a product in its own right. We plan to offer it as a SaaS for traders and investors — market analysis and price forecasting to support their own decisions when working with brokers and investment accounts.
To trade better. Our models analyse markets and forecast price moves to inform the company's own investment decisions — it's our edge, not a service we sell today.
No. Every decision is explainable: we know which signals and what probability sit behind a forecast. Interpretability is part of managing risk.
It's on our roadmap, not yet released. We're building and proving the technology on our own capital first.
On the company's own infrastructure, registered in Cyprus, in line with applicable data requirements.