Tondratex extracts patterns from market and portfolio data in real time and returns ranked, actionable recommendations. Funds stay accessible at all times — no lock-up periods, no withdrawal queues.
The system does not run on fixed reports. It ingests, scores, and re-ranks positions continuously as new data arrives.
Price feeds, volatility indices, and transaction volume data are collected and normalized on a shared timeline before any model runs.
The engine weighs historical co-movement between assets and sectors to flag exposure concentration before it becomes a liability.
Every recommendation carries a numeric risk score derived from volatility, liquidity depth, and recent drawdown behavior.
Outputs are compared against realized outcomes on a rolling basis, and model weights adjust accordingly without manual intervention.
Model outputs are recommendations, not automated trade execution. Every decision to act remains with the account holder, and all underlying data processing follows German and EU data protection requirements (DSGVO).
Traditional funds tie capital to redemption windows measured in weeks or quarters. Tondratex was built around the opposite assumption: side income should remain liquid.
Every balance shown on the dashboard reflects funds you can move immediately. There are no notice periods, no minimum holding terms, and no penalty for early withdrawal — the same policy applies whether you deposited an hour ago or six months ago.
Read the Liquidity PolicyEach recommendation passes through four distinct stages before it reaches the dashboard.
Raw market, transaction, and sentiment data is pulled from licensed feeds at fixed intervals.
Inputs are cleaned, time-aligned, and adjusted for gaps or reporting delays across sources.
Predictive models score each instrument for expected movement and assign a risk weight.
Ranked recommendations appear on the dashboard with the reasoning and risk score attached.
The same engine supports different decision types, depending on the account holder's objective.
Retail investors use ranked recommendations to shift allocation away from over-concentrated positions identified by the correlation model.
Side-hustle investors track risk scores across a small number of positions to catch drawdown risk before it compounds.
Funds held between other commitments are placed against model-ranked opportunities and withdrawn again without a holding period.
Small business operators compare projected outcomes across two or three allocation scenarios before committing operating reserves.
Users with limited time check the dashboard on their own schedule instead of tracking multiple data sources manually each day.
Every recommendation stores the data inputs and risk score used, so a decision can be reviewed after the fact.
Direct answers to the questions we receive most often from German users.
Withdrawal requests are processed without a holding period. Funds move to your linked account once the request is confirmed; there is no lock-up window at any stage of use.
Account and transaction data is processed under EU data protection law (DSGVO). Storage infrastructure and access controls follow the same regulatory baseline applied to financial platforms operating in Germany.
No. The model produces ranked recommendations with a risk score. Execution decisions remain with the account holder at every step.
Model outputs are probabilistic, not guaranteed. Recommendations include a risk score precisely so that uncertainty is visible before a decision is made, and outcomes feed back into the recalibration process.
There is no minimum holding term and no mandatory recurring deposit. You can deposit, withdraw, or pause activity at any point.
Additional questions are answered on our full FAQ page or through direct contact.
Registration takes a few minutes. Review the model's recommendations before committing any capital.
Access DashboardNo lock-up period. No recurring commitment required. Withdraw at any time.