AI Decision Engine for Active Traders
Reapgild Wealth ingests pricing, volume and volatility feeds, runs them through predictive models, and returns a ranked set of portfolio actions — rebalance, hedge, or hold. No manual spreadsheet work required.
The Logic Foundation
Manual portfolio analysis breaks down as the number of positions and data sources grows. Reapgild Wealth replaces that bottleneck with a repeatable inference loop that runs continuously, not once a day.
Price, volume, and volatility feeds are pulled from connected sources and normalised into a single structured format.
Predictive models process the normalised data against historical patterns and current market state.
Each candidate action is scored for exposure, drawdown sensitivity, and correlation to existing holdings.
The engine returns a ranked list of actions with the reasoning behind each score, not a single black-box signal.
A trader reviewing ten positions by hand can still cross-reference charts and news in reasonable time. At fifty or a hundred positions, the same process requires hours, and the data is often stale before the review finishes.
The engine does not get slower as your watchlist grows. It applies the same ingestion-to-output cycle whether you are tracking five assets or five hundred, and it logs the reasoning behind each recommendation for later review.
Core Engine
Three components work together on every cycle: a predictive model, a risk layer, and a recommendation pipeline built to scale with portfolio size.
Models are trained on historical price action and macro indicators, then refreshed on rolling windows so they reflect recent regime changes rather than stale assumptions.
Position-level exposure is recalculated whenever a connected feed crosses a threshold you configure, not on a fixed daily schedule.
The same inference pipeline handles a short watchlist or a large book without reconfiguration on your part.
One-Click Deployment
The 60-second figure refers to the time between connecting your data and receiving the first ranked set of actions, not a promise about trading outcomes.
Link a brokerage feed or import a CSV of current holdings. No manual tagging required.
Define risk tolerance, position limits, and any assets to exclude. The engine respects these as hard rules.
The engine synthesizes the data and returns ranked actions with reasoning. You approve, adjust, or dismiss each one.
Recommendation Queue — Preview
Gathering this view manually across separate platforms typically means pulling charts, cross-referencing news, and recalculating exposure by hand. The engine runs the equivalent cycle in under 60 seconds.
Methodology & Transparency
Reapgild Wealth draws on licensed market data providers for pricing, volume, and volatility history. No recommendation is generated from a single unverified source — each model run cross-checks inputs before scoring begins.
Ready When You Are
Connect your data, set your constraints, and review your first set of ranked recommendations in under 60 seconds.
No credit card required to initialize the engine.