Deployment & Integration

Sovereign by design. Integrated in weeks.

OraBrain ships as a complete stack that runs where you need it: fully air-gapped, on your premises, or as private SaaS. It plugs into the systems you already run, and your data never crosses your boundary.

No external model APIsRuns on commodity hardwareAudit trail on every decision
Deployment Models

Same model. Your choice of boundary.

Every tier runs the full OraBrain platform. The only variable is how much of the world you allow in.

Maximum isolationAir-gapped

Fully disconnected

The entire stack, intelligence included, on hardware with no route to the internet. Built for custody, defense, and sovereign requirements.

  • Zero external connectivity, ever
  • Updates delivered by signed offline packages
  • Privacy enforced by physics, not policy
On-prem

In your data center

Runs inside your infrastructure and network controls, keeping data residency and oversight fully in your hands.

  • Your hardware, your VPC, your rules
  • Connects only to sources you approve
  • Meets data-residency mandates by default
Private SaaS

Operated by Prax Labs

A dedicated, isolated environment we run for you, with the same owned-model architecture and no third-party AI in the loop.

  • Fastest path to production
  • Single-tenant isolation per client
  • Same discipline, managed for you
Architecture

Where your data lives: with you.

Most AI vendors process your data on someone else's model in someone else's cloud. Ours is a closed loop inside your boundary.

Your sources

Signals in

Payments and wires, blockchain nodes, alerts, case data, watchlists, internal policy.

Inside your perimeter

OraBrain decisioning

Agents investigate, reason, and decide on our owned LLM, running on your hardware. Nothing calls out. Nothing trains elsewhere.

Your systems

Decisions out

Verdicts, evidence files, and enforcement actions delivered to your case manager, SIEM, and Web3Firewall.

The boundary is the product. No client data leaves the environment, no external model APIs are called, and nothing you process is used to train anyone else's system.
Integration

It meets your stack where it is.

Adopting OraBrain is not a rip-and-replace. Agents attach to the systems your teams already work in, across both sides of the house.

APIs and streaming

REST APIs and event streams for submitting cases, receiving verdicts with evidence, and driving automation end to end.

Case managers and SIEM

Decisions and evidence files land in your existing case management and security tooling, so workflows do not change.

Screening and list feeds

Sanctions, PEP, and watchlist providers plug in as sources, with OraBrain doing the adjudication your rules engine cannot.

Payment rails and core banking

Wire, ACH, card, and instant-payment flows screened in-line, fitting the latency budgets of real payment operations.

Custody and signing

Web3Firewall sits at the signing layer of wallets and custody stacks, enforcing verdicts before a transaction is broadcast.

Nodes and chain data

Your own blockchain nodes and indexers feed on-chain context directly, keeping even chain analysis inside your boundary.

Implementation

From kickoff to production, in four moves.

A deployment is an engineering project with a known shape, not an open-ended consulting engagement.

01 / Connect

Connect

Stand up the stack in your chosen boundary and wire in your sources: rails, nodes, alerts, lists, and case systems.

02 / Calibrate

Calibrate

OraBrain learns your policy and your operators' judgment from inside the perimeter, so verdicts read your context, not a generic one.

03 / Govern

Govern

Run agents alongside your team, compare outcomes, tune thresholds, and lock in QA, audit, and escalation ownership.

04 / Operate

Operate

Agents take the volume. Your experts take the exceptions and the policy pen, with every decision on the record.

Governance & Documentation

Built to survive an audit.

Every decision the platform makes is designed to be examined later, by your QA team, your model risk function, or your regulator.

Governance, in the product

  • Full decision traceability: inputs, reasoning, evidence, and outcome on every case
  • Policy mapping, so each verdict ties back to the rule and risk appetite that produced it
  • Human accountability preserved: your officers own sign-off and escalation
  • Controlled change: model updates are versioned, tested, and approved by you
  • Quality assurance sampling and challenge workflows built in
Customer documentation set

Docs that ship with the platform

  • API reference
  • Connector and integration guides
  • Deployment and hardening runbooks
  • Model governance and validation pack
  • Audit and evidence export guide

Provided to customers at kickoff and maintained with every release, so your engineers and your model risk team start with answers, not tickets.

Talk to us

AI agents will transact financially. Static security will fail. Prax Labs is ready.