Taskurai

NET applications running on Azure. It is designed to eliminate the need for developers to design, host, or maintain orchestration infrastructure, allowing them to focus on building business-critical solutions rather than managing the underlying platform. The platform targets scenarios that require reliable background and batch job processing, durable business workflows, scalable workload offloading, and fault-tolerant long-running operations.

The service offers a code-first approach, enabling developers to define their background processing solutions using a YAML configuration file and deploy with a command-line interface. NET Core C using a dedicated SDK. Taskurai supports managed orchestration with customer-owned compute, where workers run within the user's own Azure subscription, helping to keep sensitive data and services private. Installation is available directly from the Azure Marketplace.

Key features include the ability to run workers on serverless containers, support for both short and long-running tasks, and seamless scaling from zero to many instances based on task volume. The platform enables building stateful applications with key-value state stores, durable checkpoints, and state management for tracking intermediate and final states. It also supports scheduling, durable sleep for suspending code over long periods, recurring Cron jobs with full time-zone support, and built-in webhooks for integrating with external systems. Distributed locks are provided to ensure consistency and reliability in sensitive operations, using a lease-based locking mechanism.

Taskurai is suitable for building new services or refactoring existing applications to offload heavy work and turn business logic into durable, reliable workflows. It is designed for developers and organizations seeking to build robust, long-running, and fault-tolerant processes on Azure with minimal operational overhead. The platform offers a local development experience without the need for local infrastructure setup, supporting both isolated task execution and local debugging of production loads. Pricing is usage-based, with options to scale to zero for no cost when idle or maintain reduced-cost workers, automatically scaling up for peak loads.