Skip to content

oryxflow documentation

Everything you need to build data-science pipelines you can believe. Declare each step as a task. The engine runs your steps in dependency order and reruns exactly what a parameter, data, or code change affects, so no result is ever built on stale data — and it records what produced every output, which makes AI data analysis reproducible by default, for humans and AI coding agents alike. There are no filenames to invent and no note to keep about which parameters produced which output; you load any result by name. None of that costs you time, either: anything already computed loads from cache instead of running again.

New here? Read Why oryxflow for the positioning, then start with Installation and the Quickstart. Already have a project that got out of hand? Go straight to Migrate a messy notebook project.

Guides for reproducible data science

  • Why oryxflow

    What it's for and when not to use it — reproducibility, lineage, and trustworthy AI data analysis, plus honest comparisons.

  • Installation

    Install oryxflow and its optional extras (cloud storage, export, dask).

  • Quickstart

    From nothing to a running, self-caching pipeline in a few minutes.

  • Transition from scripts

    Turn an existing analysis script into cached tasks.

  • Migrate a messy notebook project

    Nine notebooks and a folder of clean_v3.csv? Restructure it so a wrong number stops being possible — by hand or in one command.

  • Writing & managing tasks

    Dependencies, inputs, outputs, and save formats.

  • Workflows & Running workflows

    Wrap tasks in a flow; preview, run, and reset.

  • Parameters

    Parameter inheritance and how it drives selective reruns.

  • Task I/O formats

    Parquet, pickle, CSV, in-memory cache, and cloud storage.

  • Managing complex workflows

    Automatic code invalidation, selective resets, multi-experiment flows.

  • Experiment tracking

    How oryxflow pairs with MLflow or Weights & Biases — different halves of the same project.

  • Build with Claude Code

    Make AI-written data analysis trustworthy: scaffold the project, wire the DAG, and teach the agent to use the cache correctly.

  • Built for AI coding agents

    llms.txt for one-request ingestion, core examples executed by the test suite, and a reference generated from docstrings that cover the whole public API.

  • API Reference

    Every public symbol, generated from the source docstrings.