00 / Hero — regulatory mesh · live Monte Carlo
matches0mean EV+0.000p(profitable)0.00rake3.6%break-even— matches
IDLE
/ Observation
At skill_gap = 0.40 and rake = 3.6%, ~83% of paths close positive at T=160.
/ Reference
Each thread is one player’s equity curve. Bright line is the population mean — an ensemble of PvP paths, not a single sample.
GameRock Paper Scissors
Skill gap0.40
Rake3.6%
Paths × T1,200 × 320
Regala Labs · est. MMXXIV · regulatory OS

New rails for an old industry.

01 / Thesis

In ten years, every interesting product will cross a regulatory line. We are building the operating system for what crosses it.

Regulation was built around clean categories: securities here, gaming there, money transmission over there. The next generation of products refuses to sit on one side. Prediction markets look like both derivatives and sports betting. iGaming carries obligations from BSA/AML, state gaming, and tribal compacts at once. The single-regime answer is no longer the operator’s reality.¹

Compliance teams reconstruct the same cross-regime analysis by hand, every time a new product structure arrives — slow, expensive, and hard to defend a year later when a regulator asks the same question and the analyst who answered it has moved on. The artifact that matters is not the memo, but the chain of citations the memo rests on.²

Regala Labs exists to make that chain a first-class object. A classification simulator that answers regime by regime, anchored to statutes and case law. A player-side runtime that lives inside the same regulatory perimeter — because RG obligations and conversational-AI liability are not someone else’s problem. One operating system for operators whose products refuse a single regime.³

02 / Structure — Variance & Equilibrium
Exhibit B · interactive

A PvP simulation running live. Each thread is a player’s equity curve.

Where the histogram showed outcomes, this shows paths. Variance is visible. Equilibrium is a discipline, not a guess.

03 / Product
What we’re building

Three surfaces. One regulatory perimeter.

A back-office OS for compliance. A runtime that lives inside the same perimeter as the player. A classification engine that survives the next audit because the citations come with it.

01Back-office

Regulatory OS for operators.

For compliance teams whose products span finance, gaming, and crypto. v1 ships AI compliance for prediction-market and iGaming operators — sold to the people who answer to the regulator.

  • Cross-regime memos
  • Versioned policy registry
  • Regulator-ready exports
productregimesingestCFTCSTATEBSA/AMLSECTRIBAL
02Runtime

RG-native player surface.

Streaming-ML interventions when problem-gambling signals appear. Multilingual chat and support, safety-harnessed, multi-tenant across operators with different residency and customization requirements.

  • Real-time risk scoring
  • Operator-shaped policy
  • Multi-tenant residency
signalactionplayerintervenepolicy
03Classification

Cited. Reproducible. Defensible.

Feed in a contract or product structure. Get classification across CFTC, state gaming, BSA/AML, securities, and tribal — each anchored to statutes and case law. Same input, same output, audit after audit.

  • Statute + case-law anchors
  • Noted dissents
  • Frozen reruns
contractclassificationstructvenuepayoutregime[]cite[]dissentsame input → same output, audit after audit
04 / How it works
How it works

One pipeline. Five surfaces. Same artifact every time.

From the contract that lands in the inbox to the regulator-ready export — nothing is reconstructed by hand.

  1. 01Ingest

    Ingest the product structure.

    Contracts, market designs, payout curves, KYC posture, geofencing, payment rails. Imported as machine-readable shape — not a memo.

  2. 02Classify

    Classify across regimes.

    CFTC, state gaming, BSA/AML, securities, tribal. Each regime answers independently, with confidence and a noted dissent if it splits.

  3. 03Cite

    Generate the cited audit trail.

    Every classification ships with statutes, case law, and the version of the policy it ran against. The memo is the by-product, not the artifact.

  4. 04Monitor

    Monitor markets and player risk.

    Streaming-ML on the player surface; market-state monitors on the back-office. Drift triggers a rerun on the same input — and you can see what changed.

  5. 05Deploy

    Deploy operator-facing controls.

    Policy registry, intervention library, regulator-ready exports. The same perimeter that classified is the one that operates.

05 / Portfolio
Products

One live. One in preview.

/ betignite.us · product 001LIVEvisit ↗

BetIgnite

A skill-based PvP platform. The IGN token settles matches; the protocol publishes the rake. Provably-fair 1v1 from coin-flip-fast to full-depth strategy.

Provably fairOn-chain settlementProtocol rake · 3.6%
0
Matches settled
0
Concurrent · 24h
0.0%
Protocol rake
/ regala.ai · product 002PREVIEW

Regala

The regulatory operating system for operators crossing finance, gaming, and crypto regimes. v1 ships AI compliance for prediction-market and iGaming teams, anchored on a cross-regime classification simulator — feed it a contract, get classification across CFTC, state gaming, BSA/AML, securities, and tribal with cited statutes and case law.

Cross-regime classificationCited statutes + case lawAI compliance · v1
06 / Careers
Careers

People who would rather build a ladder than climb one.

Small founding team. Cross-regime products are written by generalists with deep specialties. Read these as starting points — if your shape isn’t listed, write anyway.

Department
Type
4 open roles
  • Market Designer

    MarketsFoundingRemote · US

    Design product structures that survive cross-regime classification. Partner with the engine to make ambiguity legible to a regulator.

    • Mechanism design for skill-based PvP and event contracts
    • Cross-regime ambiguity → product-level invariants
    • Drive the venue/payout decisions that determine regime
    Apply via Simon@regalalabs.com
  • Applied Mathematician

    ResearchFoundingRemote · US

    Probabilistic modeling for player risk and market state. Translate fairness, drift, and dissent into objects the engine can rerun.

    • Streaming-ML risk scoring under residency constraints
    • Measure-of-skill estimators with a published rake
    • Drift detection on regime classifiers
    Apply via Simon@regalalabs.com
  • Systems Engineer

    EngineeringFull-timeRemote · US

    Build the audit-trail spine: deterministic reruns, versioned policy, citation graphs that survive a year-later question.

    • Reproducible compute on the classification engine
    • Multi-tenant residency for the runtime
    • Regulator-ready exports — not screenshots
    Apply via Justin@regalalabs.com
  • Applied AI Engineer

    EngineeringFull-timeRemote · US

    Ship safety-harnessed conversational AI on a multi-tenant player surface. Liability is a product surface, not a footnote.

    • Multilingual chat with operator-shaped policy
    • Eval, redaction, and trace tooling for the runtime
    • Tight loops with the classification engine
    Apply via Justin@regalalabs.com