Easy to setup and adapt ETL tool

Its async + multiprocessing so several workers can transform data simultaneously Really helpful for Kafka or RabbitMQ

• write your own pipelines classes for feed and stream and set them in settings (oruse natively provided)
• write your own worker class or use natively provided and set it in settings.py
• write your own chain of middlewares to transform the data between sources