Barkhapragatia processes market data continuously and adjusts its recommendations to your stated risk tolerance, so your portfolio strategy keeps pace with changing conditions without requiring your daily attention.
UK markets have shown increasingly short cycles of repricing across equities, bonds, and alternative assets. For a professional managing investments alongside full-time work, reviewing this data manually on a consistent basis is rarely practical.
Spreadsheet-based tracking and periodic broker updates tend to lag behind the actual rate of change, which leaves decisions based on information that is already partially outdated by the time it is reviewed.
Most side-hustle investors can realistically check positions a few times a week, not continuously.
Risk appetite set once at account opening rarely reflects how circumstances change over time.
Price feeds, macroeconomic indicators, and sector data are rarely reviewed together in one pass.
Barkhapragatia combines predictive modelling with a continuous feedback loop, so recommendations reflect both market conditions and your own stated boundaries for acceptable drawdown.
During onboarding, you define acceptable volatility ranges and capital preservation priorities. The model treats these as constraints, not suggestions, and recalculates recommended allocations whenever market conditions shift.
Illustrative output: recommended allocation weighting across asset classes, recalculated on each analysis cycle.
The model draws on historical pricing patterns, volatility indicators, and macroeconomic datasets to produce a probability-weighted range of likely outcomes, rather than a single fixed prediction.
Illustrative output: projected return distribution bands across the analysis horizon.
Incoming data is processed as it arrives rather than batched overnight. This allows the system to flag material shifts in conditions between your scheduled review points, so recommendations stay current when you do check in.
Illustrative output: frequency of flagged data events across a rolling analysis window.
Every output can be traced back through the stages below, so recommendations are explainable rather than presented as an opaque result.
Market, pricing, and macroeconomic data are ingested from established financial data sources on a continuous basis.
The predictive model applies your calibrated risk parameters to generate a range of statistically supported scenarios.
A ranked set of allocation or positioning suggestions is produced, each annotated with its underlying rationale.
You review and approve recommendations at your own pace; nothing is executed automatically without confirmation.
Account data and analysis inputs are encrypted in transit and at rest. Access to personal financial data is restricted to the processes required to generate your recommendations, and no data is sold to third parties.
Barkhapragatia is positioned for professionals who want a disciplined, data-led second opinion on their portfolio rather than a fully autonomous trading system. Autonomous analysis handles the continuous monitoring; you retain the final decision at each review point.
This structure suits a side-hustle style of investing: engagement happens on your schedule, in short, well-informed sessions, rather than through constant manual tracking.
The same underlying model supports a small number of recurring, practical objectives for independent investors and small-scale business strategy.
Allocation suggestions are weighted across asset classes and sectors based on correlation analysis, reducing concentration in positions that move together under stress.
When volatility indicators exceed your defined threshold, the model shifts emphasis toward lower-variance instruments to limit drawdown during unsettled periods.
Within your risk parameters, the model identifies income-generating positions with a favourable risk-adjusted return profile relative to comparable alternatives.
The model is periodically backtested against historical periods it was not trained on, and its recommendations are compared against benchmark allocations to check for systematic skew toward any single asset class or sector.
Inputs include public market pricing data, published macroeconomic indicators, and sector-level financial statistics from established data providers. No non-public or insider information is used.
At onboarding, you answer a structured set of questions covering acceptable drawdown, time horizon, and liquidity needs. These responses are converted into numerical constraints that bound the model's recommendations, and can be revised at any time.
No. Barkhapragatia produces recommendations for your review. Execution requires your explicit confirmation at each review point, consistent with a low-active-involvement rather than fully automated approach.
Underlying data is processed continuously, but recommendations are only surfaced to you when a statistically significant change in conditions is detected, to avoid unnecessary noise between review sessions.
The methodology is designed for investors managing a portfolio alongside other work, where only occasional review time is available. It does not assume constant monitoring or specialist financial training.
Set your risk parameters once, and let the model handle continuous monitoring between your review sessions. You remain in control of every decision that follows.
Read the full methodology and FAQ