blueengine-docs's Introduction

BlueEngine Document

Welcome to the BlueEngine document. Whether you are a back-end developer, or an AI engineer, in order to use the BlueEngine ecosystem, there are three main parts you need to care about:

Key features

The BlueEngine SDK is what an AI engineer will use to transform an AI model, or any kind of processing code in general, to an engine that can be integrated into the BlueEdge system. Using the BlueEngine ecosystem to build engines give you several benefits:

Build an engine is as simple as writing Python functions.
Easily create waterfall style multi-step processing pipeline, each step run in its own Process.
Auto scale the Processes to handle multiple concurent inputs. This is the key feature to raise the number of CCU that the big system can handle, with minimum cost.
Easily send and receive any type of data to and from other engines or services, from JSON to image and video streams.
Create Dynamic Acyclic Graphs (DAGs) that can combine multiple engines together to form a complicated, high performance processing pipeline. DAGs can be created on web UI or with Python scripts (TODO)
Smart auto scaling mechanism, scale by number of incomming requests. Allow scale-to-zero to minimize operation cost.
Support webhooks.
Deploy to cloud, on-prem, or hybrid systems.
Ready-to-use monitoring system.