Aloke Dhonendra monitors 500+ trading pairs in real time, converting high-frequency market data into risk-adjusted recommendations for freelancers managing income between contracts.
The system processes high-fidelity data streams across multiple asset classes simultaneously, rather than tracking a narrow watchlist manually.
Price, volume and volatility data across 500+ pairs are ingested continuously, allowing the model to compare conditions across markets rather than in isolation.
The engine identifies statistical relationships between instruments, flagging when movement in one market has historically preceded movement in another.
Data streams are timestamped and weighted by recency, so decisions are based on current conditions rather than delayed snapshots.
Each stage exists to reduce the influence of noise and emotion on the outcome.
Market data from 500+ pairs is normalised into a common structure, filtering for feed quality before it reaches the modelling layer. Incomplete or anomalous data points are discarded rather than interpolated.
The aggregated data is scored against historical pattern sets to produce a probability-weighted forecast for short and medium horizons. Each forecast carries a confidence band rather than a single fixed figure.
Positions are sized according to predefined exposure limits, not solely by predicted return. This keeps individual signals from dominating the overall allocation.
On risk mitigation: Because freelance income is inherently irregular, exposure limits are set conservatively by default. The model treats capital preservation between contracts as a constraint, not an afterthought.
UK freelancers typically face variable settlement dates, staggered contracts and irregular VAT or Self Assessment liabilities. Aloke Dhonendra is built around that irregularity rather than a fixed monthly salary pattern.
Idle capital held between projects loses relative value to inflation and opportunity cost. The platform is positioned as a financial buffer: capital that would otherwise sit static is allocated according to risk-adjusted models while contracts are being sourced.
Tax-efficient growth planning is supported by exportable activity records, which can be passed to an accountant ahead of Self Assessment deadlines. The system does not provide tax advice; it produces the data an adviser needs.
Contract endsFreed-up liquidity is identified and flagged as available for allocation.
Buffer allocationCapital is distributed according to the user's set exposure limits across monitored pairs.
Ongoing monitoringPositions are reviewed continuously as new contract income and outgoings are logged.
Drawdown on demandLiquidity can be withdrawn when a new contract requires working capital.
Aloke Dhonendra was designed around a simple observation: freelancers who manage their own income rarely have the time to monitor markets manually, yet they are frequently better placed to tolerate short-term volatility than salaried workers with fixed monthly obligations.
The platform does not publish opinions on where markets are heading. It publishes the output of a defined process — aggregation, modelling and execution — and keeps a record of that process for review.
Rather than relying on testimonials, Aloke Dhonendra publishes the methodology behind its backtesting and updates it as models are revised.
Illustrative representation of directional accuracy across six historical test windows. Full backtesting parameters are available on request.
Every predictive model is versioned. Historical forecasts are retained alongside actual outcomes so accuracy can be reviewed independently of the version currently in production.
No forward-looking figure is presented as guaranteed. Backtested results describe past performance under historical conditions and do not represent future returns.
Data feeds are polled continuously and the model re-scores affected instruments within seconds of ingestion. Execution decisions incorporate a short confirmation window to avoid reacting to single erroneous ticks.
Connections use encrypted transport and scoped, revocable keys. Trading permissions and withdrawal permissions are held separately, so a compromised read-only key cannot authorise fund movement.
Exposure limits are recalculated when volatility exceeds historical thresholds for a given pair. In practice this typically reduces position sizing or pauses new entries on affected instruments until conditions normalise.
Yes. A demo mode replays the same data and modelling pipeline against a simulated balance, so the decision process can be reviewed before any live capital is allocated.
It is not designed for users seeking guaranteed short-term returns or those unwilling to accept any capital risk. It is built for users comfortable with data-led, risk-adjusted allocation over time.
Set-up takes a few minutes, and demo mode allows the full decision process to be reviewed before any live capital is committed.
Create an AccountNo obligation to fund the account immediately. Demo mode is available on sign-up.