Open Source · IC/UFF e-Science Research Group
One Workflow.
Multiple Platforms.
Define a scientific workflow, choose where it runs, make a plan, execute it, and inspect the result. Start locally; connected environments require their own setup.
Native client for macOS, Windows, and Linux, distributed through versioned GitHub Releases.
Choose downloadInstallation guideDesktop starts the local AkôFlow service through Docker · Full installation guide
Explore the documentation
Browse by purpose. These sections match the documentation sidebar.
Tutorials
Install Desktop, make a first run, connect infrastructure, and follow complete examples.
How-to guides
Task-focused instructions for workflows, environments, artifacts, and operations.
Explanations
Understand planning, runtimes, network estimates, and the evidence left by a run.
Developing AkôFlow
Explore the engine, runtime adapters, and module boundaries.
Reference
Find API endpoints, payloads, environment formats, and execution states.
Contributing
Improve the documentation and keep examples aligned with verified behavior.
Example workflows
Preview every workflow graph by section. Select a card to open its walkthrough or pattern.
Simple examples
Simulation, real execution, and adapter fixtures. Each page states its prerequisites and validation scope.
Machine Learning
Training, tuning, inference, evaluation, and model delivery patterns. These are DAG designs, not verified executable bundles.
Generative AI
Data preparation and evaluation patterns for retrieval, generation, audio, and documents.
Agentic workflows
Bounded agent execution with explicit review, policy, and evidence stages.
Scientific AI
Simulation, analysis, surrogate models, and experiment feedback loops.