Temporal

Temporal is an open-source workflow orchestration platform that enables engineers to write long-running, failure-resistant business logic as ordinary code — without managing queues, retries, state persistence, or distributed system failures manually. Founded in 2019 by former Uber engineers who built Cadence (Uber's internal workflow engine), Temporal is used at companies like Netflix, Stripe, Coinbase, Box, and Datadog for mission-critical business processes. Temporal's core abstraction is the Workflow function — a function written in your application code (Go, Python, Java, TypeScript, PHP, Ruby, or .NET) that can run for seconds, minutes, days, or years. Temporal automatically persists the workflow's execution state (Event History) at every step, enabling any workflow to resume exactly where it left off after any failure — network interruption, server crash, or deployment. This eliminates the "compensating transaction" patterns and complex state machine code typically needed for reliable distributed processes. Activity functions within a workflow execute the actual work (API calls, database operations, file processing) with configurable retry policies, heartbeat timeouts, and schedule-to-close deadlines. Temporal's query and signal APIs allow external systems to interrogate workflow state or send events to running workflows without polling or additional state storage.