Rebound

Navigation

Explore

Choose a platform module or vertical, then jump straight to the corresponding page.

How it worksGoPricingGoFAQGo
Verticals / Quantitative Finance

FinAlgo
model risk, governed in real time

A quant fund's alpha models are always decaying under regime change. Standard practice: retrain often, add human override, hope. FinAlgo replaces hope with a mathematically bounded trust horizon — and refuses to trade when forecasts cross the noise floor.

18–34%improvement in decision utility across domains when horizon-constrained selection replaces accuracy-based selection.
The pain point

A single bad model decision in a volatile, Lyapunov-high regime can wipe out months of gains. No existing tool refuses to sign an execution signal because the forecast has entered noise territory.

Use case

A systematic stat-arb desk running 200+ strategies across equities, FX, and rates.

Without Rebound
  • One 'best' model picked by validation RMSE, deployed across all horizons.
  • Drift detected after P&L damage has already accrued.
  • Risk officers reconstruct model-selection rationale by hand for FINRA SR 11-7 exams.
  • Execution continues blindly through regime transitions.
With Rebound
  • A Rashomon pool: transformer, reservoir, CNN-LSTM, GBDT — governed as one object.
  • Per-strategy k* computed live; signals beyond it are blocked before execution.
  • Model choice rationale and rejected alternatives auto-logged in a signed audit ledger.
  • Regime transitions trigger automatic re-routing to the safest model for the new regime.
Concrete example

stat-arb-v7, 5-minute horizon, chaotic regime — a real-shape trace.

finalgo.Rebound.io / strategies / stat-arb-v7 / liveSignal stream · NYSE · 1s tick
Strategy
stat-arb-v7
USD · 48 instruments
Current k*
7
minutes ahead · regime: chaotic
Last hour
41signals cleared
6blocked · k > k*
Most recent signal
POST /v1/select
{
  "pool": "stat-arb-v7",
  "horizon_min": 5,
  "risk": "risk_adjusted_return",
  "instrument": "XYZ"
}

200 OK
{
  "decision": "ALLOW",
  "model":    "reservoir-net-fx",
  "utility":  0.91,
  "confidence_envelope": [ -0.0023, 0.0041 ],
  "k_star":   7,
  "regime":   "chaotic",
  "audit_id": "ax_fa_2f9e81c3",
  "sign":     "sha256:<truncated>…"
}
Ideal customer profile
  • Hedge funds with AUM ≥ $500M running systematic or quant strategies
  • Prop trading desks at Tier-1 investment banks
  • Algorithmic market makers
  • Risk and model-risk departments at asset managers
Buyers we work with
CTOHead of Model RiskChief Risk OfficerHead of Systematic Trading
Typical deployment
Pulse API
API call volume + model count
$8K–$25K / mo
Command Platform
AUM-tiered, multi-strategy
$80K–$250K / yr
Performance Share
Value-aligned with gains delivered
Base + 5–8 bps

Ready to scope a pilot?

A 60-day Predictability Audit on your real workload, no commercial commitment.

Talk to usNext: CodeSentinel