Orenyx AI Engine™
Skip to content
Log inStatus
Orenyx AI Engine™
HomeAboutFeaturesPricingUse CasesDevelopersIntegrationsSecurity & Trust
Get Started
Orenyx AI Engine™
Orenyx AI Engine™

Orenyx AI Engine™ — unified decision plus orchestration layer for dispatch, bots, and payments.

Product

  • Features
  • Pricing
  • Use Cases
  • Integrations
  • Security

Developers

  • API Docs
  • Authentication
  • Changelog
  • Status

Company

  • About
  • Contact
  • Careers
  • FAQ

Legal

  • Terms of Service
  • Privacy Policy
  • Refund & Cancellation Policy
  • Data Processing Agreement

© 2026 Orenyx Labs LLC. All rights reserved.

FAQ

The questions we get asked most. More will be added as they come in.

  1. Home
  2. /
  3. FAQ

A unified decision and orchestration layer for dispatch, bots, and payments — one engine that every product in your stack shares instead of each reinventing its own logic.

[PLACEHOLDER — answer for “How is this different from building automation in-house?”]

[PLACEHOLDER — answer for “How is this different from Zapier/Make or a generic workflow tool?”]

No. The engine powers the Orenyx ecosystem, but it is available as a standalone platform and connects to your own systems through the API and webhooks.

[PLACEHOLDER — answer for “What does “payment decisioning” actually mean, and who is liable for the decision?”]

Strict per-tenant isolation, with no cross-tenant data access. Data is encrypted in transit (TLS 1.2+) and at rest (AES-256), and sandbox and production environments hold isolated data with no cross-environment leakage.

Usage above your plan’s included allowance is billed as overage: $2.00 per 1,000 events on Starter and $1.50 per 1,000 on Growth. Enterprise overage is negotiated. Nothing stops working when you cross the line.

Every account includes a sandbox environment with test API keys. Sandbox events do not incur usage charges and are clearly flagged in logs.

[PLACEHOLDER — answer for “How long does integration typically take?”]

Starter includes email support with a 48-hour response target. Growth adds chat and a 24-hour target. Enterprise is dedicated support, SLA-backed.

Still have a question?

Ask us directly and we will get back to you.

Contact us →