Predictive risk floorSystem 01 / 10

QUANTPROP // VERIFIED CASE STUDY

Real-Time Predictive Risk Engine

Predictive edge. Controlled risk.An event-driven sports analytics and risk platform with versioned policy evaluation, immutable alert evidence, exact wager capture, append-only settlement, leakage-safe replay, CLV measurement, and regression watchdogs.
AI CAPABILITY PLANEProbability, governance, and evidence operating as one controlled decision loop
Market + StatcastVerified inputs
Feature blendingForm + context
Simulation enginesMonte Carlo + NB
Risk gatesEdge + watchdogs
RecommendationHuman execution
Verified
Simulation intelligence

Runs probabilistic game and player-outcome scenarios.

Market + featuresProbability + edge
Verified
Model governance

Tracks calibration, cohorts, regression, and promotion status.

Model telemetryGate decision
Verified
Evidence generation

Preserves versioned alerts, accepted prices, settlement, and CLV.

Decision stateAuditable record
01 // SYSTEM TOPOLOGY

Connected architecture

Event-driven decision platform spanning source acquisition, feature computation, probabilistic scoring, governed recommendation delivery, settlement, and closed-loop model monitoring.

PlatformData planeEdge / externalAuthority / control
01 // Source acquisition
Market feedsSportsbook lines + price movement

WHYEstablishes trusted analytical inputs

Verified
MLB / StatcastPlayer, game + pitch features

WHYEstablishes trusted analytical inputs

Verified
Weather contextVenue + game conditions

WHYEstablishes trusted analytical inputs

Verified
02 // Orchestration
Airflow control plane24 scheduled and event-driven DAGs

WHYMakes pipeline execution repeatable

VerifiedCONTROL GATE
Quality contractsFreshness, schema + cohort checks

WHYMakes pipeline execution repeatable

VerifiedCONTROL GATE
03 // Operational data
TimescaleDB27-table analytical system of record

WHYStores current and historical decision state

Verified
RedisHot features + execution state

WHYStores current and historical decision state

Verified
04 // Model intelligence
Feature blendingForm, matchup, park + market signals

WHYTurns features into probability estimates

Verified
Probability enginesNegative binomial + Monte Carlo

WHYTurns features into probability estimates

Verified
05 // Decision governance
Edge + tier policyGold, Silver + Bronze contracts

WHYConstrains recommendations and authority

Verified
Circuit breakersRegression and hit-rate gates

WHYConstrains recommendations and authority

VerifiedCONTROL GATE
Focus Contract v2Versioned pure policy evaluator

WHYConstrains recommendations and authority

Verified
Alert evidence ledgerImmutable runs, chunks + picks

WHYConstrains recommendations and authority

Verified
06 // Measurement loop
Exact wager captureAccepted line, price, stake + time

WHYMeasures outcomes against recommendations

Verified
Settlement ledgerAppend-only exact-game revisions

WHYMeasures outcomes against recommendations

Verified
Policy replayRevision-pinned leakage-safe evaluation

WHYMeasures outcomes against recommendations

Verified
CLV analyticsPrice-quality measurement

WHYMeasures outcomes against recommendations

Verified
07 // Observability
Grafana20-panel operational evidence

WHYMakes system health and outcomes visible

Verified
PrometheusPipeline and service health

WHYMakes system health and outcomes visible

Verified
SupersetAnalytical exploration

WHYMakes system health and outcomes visible

Verified
08 // Feedback + authority
Model watchdogsCalibration, cohort + promotion gates

WHYKeeps final authority with operators

VerifiedCONTROL GATE
Operator controlAlerting + manual execution authority

WHYKeeps final authority with operators

VerifiedHUMAN AUTHORITY
GOVERNED OUTCOMEFocus Contract v2 is deployed across the lab cluster with two-sided paper recommendations, immutable delivery evidence, closing snapshots, settlement, policy replay, and a completed post-deploy acceptance run; automatic wagering remains disabled.Ingest → Evaluate → Evidence → Capture → Settle → Replay → Gate
QUANTPROPEVIDENCE CHAIN
  1. 1Market SnapshotVerified
  2. 2Focus Contract v2Verified
  3. 3Alert EvidenceVerified
  4. 4Wager CaptureVerified
  5. 5Closing EvidenceVerified
  6. 6SettlementVerified
  7. 7Policy ReplayVerified
253 TESTS • 7 EVIDENCE MIGRATIONS • DEPLOYMENT ACCEPTED
AUTOMATIC WAGERING DISABLED

PURPOSEA recruiter-readable visual summary of the detailed topology above; verification labels preserve the boundary between demonstrated, pending, and historical evidence.

02 // PROOF LIBRARY

Evidence gallery

Live model track record
Claim state: VerifiedArtifact type: Verified runtime captureReviewed: 2026-07-27
Live model track record

Read-only Grafana evidence showing 3,145 graded recommendations, realized win rate, outcome distribution, and performance segmented by tier and market over the selected 30-day view.

Open full evidence QuantProp private operational Grafana dashboard; captured through a temporary least-privilege Viewer account and sanitized of infrastructure identifiers
Daily performance telemetry
Claim state: VerifiedArtifact type: Verified runtime captureReviewed: 2026-07-27
Daily performance telemetry

Operational trend view exposing day-level variance instead of presenting only aggregate outcomes; adjacent tables retain market and tier context.

Open full evidence QuantProp private operational Grafana dashboard; read-only Viewer capture with infrastructure and account details excluded
Airflow orchestration inventory
Claim state: VerifiedArtifact type: Verified runtime captureReviewed: 2026-07-27
Airflow orchestration inventory

Live read-only inventory of the QuantProp DAG family, including feature computation, calibration, CLV, contract-cohort monitoring, and daily decision workflows.

Open full evidence Private Airflow control plane; temporary Viewer account with account identity redacted
MLflow promotion-gate history
Claim state: VerifiedArtifact type: Verified runtime captureReviewed: 2026-07-27
MLflow promotion-gate history

Forty tracked model-validation and promotion-gate runs demonstrate an auditable evaluation history rather than an unversioned model claim.

Open full evidence Private MLflow experiment tracker; host and infrastructure identifiers excluded
Orchestration control plane
Claim state: VerifiedArtifact type: Repository-derived evidenceReviewed: 2026-07-27
Orchestration control plane

Sanitized repository facts spanning 24 DAG files, ingestion, model controls, settlement, CLV, and watchdogs.

Open full evidence Local QuantProp DAG source and SYSTEM.md; not live telemetry
Model promotion gates
Claim state: VerifiedArtifact type: Repository-derived evidenceReviewed: 2026-07-27
Model promotion gates

Public rendering of verified shadow, calibration, cohort-health, warning, alert, and manual-authority logic.

Open full evidence Local QuantProp watchdog source; synthetic presentation of verified logic
Architecture mapAvailable now

Sanitized logical topology with evidence states.

Dashboard / DAGAvailable now

Sanitized repository-derived dashboard or orchestration evidence is included above.

Video / demo clipCapture pending

A short captioned product proof is still required.

Product screenshotAvailable now

Reviewed product-interface evidence is included above.

NEXT CONVERSATIONDiscuss this architecture

Explore how this delivery pattern can be adapted to a governed enterprise deployment.

PRIVATE REVIEW ROOMSensitive evidence stays private

Production endpoints and restricted artifacts remain disconnected until an identity provider and access policy are explicitly configured.