- What is m8?
- Why m8flow?
- Integration
- Watch and Interact
- How to get started
Anchored by m8 Principles
The m8 Philisophy
- merge flows effectively: human tasks, DMN rules, and AI agents in one BPMN pipeline.
- make apps faster: build on Django, Flask, and FastAPI instead of a JVM sidecar.
- manage processes better: multi-tenant by design, one engine for every team.
- minimize errors: deterministic execution and full audit trails instead of brittle scripts.
- maximize efficiency: BPMN and DMN orchestration on a fraction of the infrastructure.
- model workflows visually: BPMN 2.0 diagrams your whole team can read.
- modernize systems: replace JVM engines without rewriting what works.
- mobilize innovation: plug AI agents in through MCP, like adding an API.
Why m8flow instead of traditional workflow engines?
Resource Light Engine
Major BPMN engines like Camunda, jBPM, and Flowable are Java-based and typically require substantial CPU, heap, and disk space, even for modest workloads. m8flow is built on SpiffWorkflow and designed to be resource-light, delivering full BPMN 2.0 and DMN orchestration without the bloated infrastructure or cloud costs of a JVM.
Native to Your Python Ecosystem
Modern data, AI, and web development run on Python. Traditional Java-based engines force you to cross runtime boundaries with complex middleware, REST APIs, or external workers. m8flow runs natively in pure Python: your orchestration logic can securely and directly invoke your existing Django, Flask, and FastAPI services, and your ML libraries, without a JVM sidecar.
Predictable Costs & Truly Open Source
Even open-source Java engines carry complex runtime licensing, like Oracle’s employee-based Java SE pricing, that can produce unpredictable cost spikes. m8flow eliminates JVM procurement and licensing risk entirely. Released under Apache 2.0, it’s a scalable, multi-tenant alternative with zero vendor lock-in and no hidden maintenance fees.
Architected for an AI-Driven Future
Traditional BPMN engines are proven in regulated environments because they emphasize audit trails and deterministic execution, but they struggle to adapt to modern AI. Legacy platforms bolt on AI through external modules and brittle connectors. m8flow is built for agent pluggability from the ground up, with native support for the Model Context Protocol (MCP). Human tasks, deterministic DMN rules, and autonomous AI agents run in the same governed, auditable pipeline, so you get next-generation AI capability without giving up the controls a regulated enterprise depends on.
One engine,
one platform.
m8flow doesn't run alone. It's the orchestration core behind every product in the AOT suite.
formsflow.ai is where people meet the process. A form can start a workflow in m8flow, feed it, or wait on it. The person filling it in and the process running behind them never drift apart.
Unstructured case management for work that doesn’t fit a rigid flowchart. caseflow hands off structured sub-processes to m8flow’s BPMN engine, so case workers get flexibility where they need it and governed process where it matters.
Connector management for everything m8flow needs to talk to. node-wire handles the integrations, credentials, and connector lifecycle, so your workflows can reach external systems without hardcoding a connection for every one.
The public sector and higher-ed solution stack, built on the same core. SLED360 packages m8flow’s orchestration with the compliance and workflow patterns state, local government, and education institutions actually need out of the box.
Example Headline
Can Go Here
With a Secondary Headline
Video Popup
Interactive Popup
How to use/deploy
Ready to
build on m8flow?
Open source, Apache 2.0, and backed by AOT’s implementation and support services when you need them.
Need implementation, training, or support? Our team can help. Public sector or higher ed? See how m8flow fits into SLED360.
