Architects resilient, durable execution workflows for AI agents and production-grade automation systems.
This skill transforms fragile scripts into production-ready infrastructure by implementing durable execution patterns that ensure workflows resume exactly where they left off during failures. It provides specialized guidance on selecting the right platform tradeoffs—between the accessibility of n8n, the strict correctness of Temporal, and the developer experience of Inngest. By leveraging sequential, parallel, and orchestrator-worker patterns, it helps developers build reliable systems for payment processing, background jobs, and complex AI agent interactions while avoiding common pitfalls like side effects and lack of observability.
Key Features
01Best practices for idempotency and exponential backoff
02Durable execution for resilient multi-step processes
030 GitHub stars
04Implementation of sequential, parallel, and orchestrator-worker patterns
05Platform-specific optimization for n8n, Temporal, and Inngest
06Advanced error handling and failure recovery strategies
Use Cases
01Building multi-step payment flows that require high reliability and recovery
02Migrating brittle cron jobs to a managed durable execution framework
03Orchestrating complex AI agent tasks across distributed worker nodes
