Pipes
Pipes
Built-in visual workflow engine for event processing, AI pipelines, and automation
Pipes is npayload's built-in workflow engine that runs inside every instance. Build n8n-style automation workflows that process events, call APIs, run autonomous workflows, and orchestrate complex business logic, all without leaving your npayload environment.
Why Pipes exists
Every messaging system eventually needs to do something with messages. With other platforms, you bolt on separate tools: Zapier for automation, n8n for workflows, custom Lambda functions for processing. With npayload, Pipes handles all of this natively.
What makes Pipes different
| Feature | Pipes | External tools (n8n, Zapier) |
|---|---|---|
| Data access | Sub-millisecond (co-located) | Network round-trips |
| Transactions | Atomic (update data + publish events) | No atomicity |
| Security | Same auth, same encryption | Separate credentials |
| AI nodes | First-class workflow nodes | Plugin ecosystem |
| Durability | Checkpoint/resume built in | Varies by tool |
| Connectors | 93 built-in across 15 categories | Varies |
| Deployment | Zero setup (built in) | Separate infrastructure |
Key capabilities
- 15+ node types: triggers, logic, actions, and AI nodes
- 93 connectors: databases, payments, communication, storage, CRM, and more
- Durable execution: workflows checkpoint state and resume after failures
- AI nodes: classify, extract, generate, and autonomous agent nodes
- Saga orchestration: multi-step workflows with automatic compensation
- Atomic transactions: update data and publish events in a single operation
- Marketplace integration: publish workflow templates for your subscribers
For any application
Pipes is not limited to data providers. Any application built on npayload can use Pipes:
- E-commerce: Order processing, inventory sync, fulfillment automation
- SaaS platforms: User onboarding flows, billing events, usage tracking
- Fintech: Payment processing, compliance checks, fraud detection
- AI systems: Multi-model pipelines, content moderation, data enrichment
- IoT: Device event processing, alerting, telemetry aggregation
Get started
Next steps
- Building workflows for step-by-step workflow construction
- Execution model for durability, checkpointing, and failure handling
- Saga workflows for multi-step orchestration with compensation
Was this page helpful?