AI Developer Resources
Build intelligent enterprise applications with Xperiflow
Welcome to the AI Developer Resources – your comprehensive guide to building AI-powered solutions on the OneStream platform. Whether you're integrating language models into financial workflows, automating complex data pipelines, or creating intelligent decision systems, this documentation will guide you from your first API call.
Who This Documentation Is For
This section is designed for developers who want to:
-
Integrate AI capabilities into OneStream applications using Xperiflow Business Rules (XBR)
-
Automate workflows with intelligent routines and job orchestration
-
Build data pipelines that leverage machine learning and language models
-
Create enterprise-grade solutions that scale with organizational needs
We assume familiarity with C# and the OneStream platform fundamentals.
What You'll Learn
Getting Started
Begin your journey with the foundations. Understand what Xperiflow is, how XBR fits into the ecosystem, and set up your development environment in minutes.
Concepts
Master the building blocks of Xperiflow development. Learn how routines work, orchestrate complex workflows with Conduit, manage storage, and understand the security model.
Guides
See it all come together. Walk through real-world examples and learn patterns used in production deployments.
Examples
Explore bite-size code examples to see simple examples of performing actions with XBR.
The XBR Advantage
Xperiflow Business Rules (XBR) is your gateway to AI-powered development. With a single unified API, you gain access to:
Capability | Description |
|---|---|
Routines | Execute versioned, reproducible AI workflows with artifact management |
Conduit | Orchestrate complex jobs with scheduling, monitoring, and parameter management |
MetaFileSystem | Persistent storage operations integrated with the Xperiflow ecosystem |
ETL | Extract, transform, and load data with built-in tabular and unstructured data support |
Language Models | Interact with GPT-4o, o1, and other frontier models through a simple, consistent interface |
// Your first XBR code — it's this simple
var routineClient = XBRApi.Routines.GetRoutineClient(si);
var instance = routineClient.CreateRoutineInstanceAsync("MLRegessor", "1.0.0").Result;
var ctorRun = instance.CreateConstructorRunAsync(inputParams).Result;
var trainRun = instance.CreateMethodRunAsync("train", inputParams).Result;
How This Documentation Is Organized
We've structured these resources to support different learning styles and use cases:
Learn – Start with Getting Started for a guided introduction. Progress through Concepts to build deep understanding.
Build – Jump into Guides and Examples when you're ready to implement. Each guide includes complete, runnable code.
Reference – Navigate the Xperiflow Business Rules and Xperiflow Routines Documentation when you need precise details on methods, parameters, and return types.
Design Philosophy
The XBR library is built on principles we believe make for great developer experiences:
Discoverability – The XBRApi static class is your single entry point. Every capability is accessible through intuitive, namespaced properties like XBRApi.Routines, XBRApi.Conduit, and XBRApi.Etl.
Consistency – All async operations follow the same patterns. All clients are created through factory methods. Error handling is uniform across the library.
Composability – Each subsystem works independently but integrates seamlessly. Combine routines with ETL, pipe results to storage, orchestrate everything with Conduit.
Safety – Thread-safe by design. Comprehensive error handling with meaningful exceptions. Version management that prevents breaking changes.
Ready to Begin?
New to Xperiflow?
Start with What is Xperiflow? to understand the engine and how XBR fits in.
Ready to code?
Jump to Developer Setup to configure your environment in minutes.
Have experience?
Explore Concepts to deepen your understanding or browse Guides for implementation patterns.
Help us improve this documentation
Found an issue? Have a suggestion? We're committed to making these resources world-class. Open an issue or contribute directly.
A Note on This Section
The AI Developer Resources section is actively evolving. We're continuously expanding coverage, adding examples, and refining explanations based on developer feedback.