Mathematical edge, engineered for live capital.
Most quants stop at the backtest. Most developers can't read the math. We close that gap: an ELO-ranked strategy swarm now executes real capital through institutional pipelines on Google Cloud — seven-gate pre-trade risk, HMM regime awareness, and sub-millisecond hot-path execution between every signal and the market.
Engineered precision, measured where it counts.
These are the system characteristics our live pipeline is engineered and monitored against — the filtering stack, the risk envelope, and the execution path. Verified figures are shared under NDA through institutional access.
Signal-precision and risk-envelope figures describe engineered system characteristics and internal monitoring targets, not promised returns. Trading involves substantial risk of loss. Past performance, simulated or live, is not indicative of future results.
Bridging the quant–dev gap.
Research notebooks don't route orders. Crash-tested infrastructure doesn't invent alpha. Value is created in the conversion — and that conversion is our core competence.
From theory to fault tolerance
A mathematical edge is only real if it survives the bad day: the feed gap, the broker disconnect, the volatility spike. We convert theoretical alpha into fault-tolerant, high-performance live trading infrastructure — because ours has to survive too.
Institutional execution pipelines
Optimal TWAP/VWAP execution, high-throughput order routing, broker gateway integration, and idempotent async queues — the same execution architecture we run our own capital through, built on Cloud Run, Spot VM batch parallelism, and a Redis hot tier.
Verified operational alpha
We don't sell backtests. We market outcomes and operational robustness: deflated Sharpe under multiple-testing correction, max drawdown containment, and regime-segmented performance — measured on live capital, not cherry-picked paper windows.
One swarm. Four load-bearing agents.
A distributed multi-agent ecosystem where strategy genomes earn capital through tournament play, a warden gates every order, and an HMM watches the macro regime in real time. Hybrid C# (.NET 8/9) and Python microservices on GCP.
Evolution & Tournament Engine
Genome pools compete in ELO-rated tournaments on identical market paths. Surrogate models pre-score candidates; genetic mutation pipelines breed the survivors. Capital allocation follows measured tournament rank, not opinion.
Risk Warden & Circuit Breakers
Seven pre-trade execution checks between every signal and the broker: position and gross exposure floors, dynamic drawdown thresholds, real-time position sizing, and automated kill-switch fail-safes that flatten the book without a human in the loop.
Execution Layer
Optimal TWAP/VWAP slicing, high-throughput order routing, and resilient asynchronous ingestion queues integrated with broker gateways. Slippage is treated as a modeled cost, not an accident.
Regime Adaptation
Multi-state Hidden Markov Models classify the macro regime — Bull, Bear, Neutral, Macro Volatility — in real time. Position sizing, strategy weighting, and exposure caps all key off the posterior, not a static schedule.
Eight genome-optimized strategies, live.
Every strategy below runs as an evolving genome — parameters are mutated, tournament-tested, and ELO-ranked by the evolution engine before they're trusted with capital. Thresholds are elastic; nothing is hand-frozen.
Dynamic RSI Snap-Back
Regime-conditioned overbought/oversold oscillators with adaptive threshold elasticity — snap-back entries only when the HMM posterior agrees.
Dual EMA Crossover
Trend-following momentum with dynamic filter bands that widen in volatility and tighten in quiet tapes.
MACD Volatility-Scaled
Adaptive signal-line velocity and histogram divergence, scaled by realized volatility so the same genome breathes with the market.
ADX Trend Expansion
Regime-gated directional index tuned for explosive trend capture — stands down when expansion conditions aren't met.
Volume Breakout Spike
Liquidity surge detection coupled with institutional footprint tracking — the genome watches who else is in the name.
Dynamic Bollinger Mean-Reversion
Volatility-envelope reversion with multi-tier take-profit ladders, sized against live drawdown budgets.
Stochastic Momentum Flow
Fast/slow %K/%D momentum cycles calibrated across multi-timeframe panels for cycle-turn timing.
TMFC Composite (Trend–Momentum–Flow)
Multi-factor composite fusing price action, momentum, and cross-asset flow signals into a single conviction score.
Why teams choose us over the alternatives.
| Full-time quant hire | Generalist consultancy | 1.21 Initiative | |
|---|---|---|---|
| Cost structure | £150k–£250k/yr, ongoing | Open-ended billable hours | Fixed-scope engagements with defined deliverables |
| Skin in the game | Salary either way | None — incentivised by hours | Live capital behind the same architecture we build for you |
| Who does the work | Depends who you can attract | Partners sell, juniors deliver | The person you talk to is the person who builds |
| Speed | 3–6 months to hire, more to ramp | Weeks of discovery before code | AI-accelerated delivery — working software in weeks |
| What you keep | Knowledge walks when they do | Dependency by design | Full code ownership, docs, and handover — zero lock-in |
Allocators ask for verified numbers. Builders ask for architecture. Pick your door.
Prospective fund allocators and B2B clients can request gated institutional access — verified metrics, regime matrices, and live dashboards under NDA. Engineering teams can book a free 30-minute working session instead. Same-day response, no sales deck.
Access is qualification-based and not guaranteed. Nothing on this site is investment advice, an offer to sell, or a solicitation of an offer to buy any security.