Introducing AI to the Flowable Process Engines
summary
Flowable runs a BPMN loan process with REST-enabled forms and DMN decision tables, logs asynchronous history to a message queue, streams that JSON via RabbitMQ → Elasticsearch, and kicks off a Spark decision tree job that trains on about ~300k generated instances to predict accept/reject outcomes. The demo builds prediction-based UI hints and auto-rewrites a user review task into an automated DMN task; the pipeline executes end-to-end and yields actionable rules.
AI-written summary · about
chapters (12)
- 0:00What is Flowable?
- 3:06Why Flowable?
- 20:51Truly Adaptive and Dynamic BPM
- 23:03Machine Learning The BPM lifecycle
- 24:12Where can ML help?
- 26:38Demo architecture
- 29:11Embedded Process Engines
- 30:11Process Service
- 36:49Listener Service
- 39:28Part 3: Decision Analysis Service
- 39:47Part 3: Phases of ML
- 52:05Remember...
related talks
50mIntroducing the Flowable Process Engines
01h 03mOne Flow to Rule Them All: Building the end to end ML project to production
01h 29mFull-stack Machine Learning and Generative AI for Data Scientists
31mBuilding Agentic AI Platforms
38mAI made easy with Flink + AIFlow
38meXplainable Predictive Decisioning
16mOperationalizing AI in Finance: Real-Time Analytics
41mAdding Generative AI to Real Time Streaming Pipelines