JobFlow
JobFlow is a high-performance distributed task processing platform designed as a C++20-based alternative to Celery or Sidekiq. Built on a microservices architecture, it leverages userver, stackful coroutines, PostgreSQL, and Redis to enable the scalable execution of heavy background operations independent of the primary API. The system features an asynchronous Task API, a dedicated scheduling mechanism, and a network of parallel workers. Users submit tasks via REST endpoints, triggering a lifecycle that includes validation and persistence to PostgreSQL, detection by a scheduler that queries for ready tasks, dispatch to a low-latency Redis queue, and parallel execution by scalable worker nodes that update task states upon completion or failure. The project includes a comprehensive suite of unit and integration tests using Google Test and Pytest. Key technical components include an API layer for asynchronous HTTP handling, persistence layers for ACID-compliant metadata storage and high-speed message brokering,