Skip to content

docs: Add blog post on unified transformation framework #5858

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Draft
Copilot wants to merge 4 commits into
base: master
Choose a base branch
from
Draft

docs: Add blog post on unified transformation framework #5858

Copilot wants to merge 4 commits into from
+494 −0

Conversation

Copy link
Contributor

Copilot AI commented Jan 14, 2026 •
edited
Loading

What this PR does / why we need it:

Adds comprehensive blog post documenting Feast's unified transformation framework, covering the Transformation API, multiple compute engines, and migration patterns.

Changes

Blog Post (infra/website/docs/blog/transformation-framework.md)

  • Documents unified Transformation API with 6 execution modes (Python, Pandas, Spark, Spark SQL, Ray, SQL, Substrait)
  • Explains ComputeEngine architecture (LocalComputeEngine, SparkComputeEngine) with DAG-based execution
  • Includes mermaid diagrams for architecture and feature build flow
  • Provides migration guide from legacy On Demand Feature Views
  • Contains real-world use cases and code examples
  • Authors: @HaoXuAI and GitHub Copilot

Documentation Updates (docs/getting-started/architecture/feature-transformation.md)

  • Added callout linking to blog post
  • Documented available TransformationMode options with use case guidance

Example Usage

from feast.transformation.base import transformation
from feast.transformation.mode import TransformationMode

# Define transformation with decorator
@transformation(mode=TransformationMode.SPARK)
def transform_features(df: DataFrame) -> DataFrame:
    return df.withColumn("scaled", df["feature"] * 2)

# Apply to FeatureView
feature_view = FeatureView(
    name="driver_stats",
    feature_transformation=transform_features,
    schema=[...],
    source=data_source,
)

Which issue(s) this PR fixes:

Addresses documentation request for transformation framework work from PRs #5181, #5223, #5278, #5185, #5281, #5287, #5354, #5445, #5521, #5523, #5666, #5611, #5612, #5502, #5650, #5629

Misc

Blog post covers architectural improvements that unify feature transformation logic across compute engines while maintaining backward compatibility with existing APIs.

Original prompt

This section details on the original issue you should resolve

<issue_title>Add blog post on new transformation framework</issue_title>
<issue_description>Is your feature request related to a problem? Please describe.
I want a blog post on the website that outlines the new transformation framework in feast. Please see these PRs:

Describe the solution you'd like
Please make sure the blog post is under infra/website/docs/blog and it would probably good to update /docs if needed.

Feel free to add a mermaid diagram or so. Please list @HaoXuAI and GitHub Copillot as the authors.
</issue_description>

<agent_instructions>I want a blog post on the website that outlines the new transformation framework in feast. Please see these PRs:

Please make sure the blog post is under infra/website/docs/blog and it would probably good to update /docs if needed. Also, please make sure to look at the existing documentation before updating.

Feel free to add a mermaid diagram or so. Please list @HaoXuAI and GitHub Copillot as the authors.
</agent_instructions>

Comments on the Issue (you are @copilot in this section)

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

Copilot AI changed the title [WIP] Add blog post on new transformation framework docs: Add blog post on unified transformation framework Jan 14, 2026
franciscojavierarceo
franciscojavierarceo reviewed Jan 14, 2026
View reviewed changes
docs/getting-started/architecture/feature-transformation.md
returns some set of output data. Feature transformations can happen on either raw data or derived data.

{% hint style="info" %}
**New in Feast**: Check out the [Unified Transformation Framework blog post](../../blog/transformation-framework.md) to learn about the latest enhancements to feature transformations, including support for multiple compute engines and a consistent API across different transformation modes.
Copy link
Member

@franciscojavierarceo franciscojavierarceo Jan 14, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@copilot this will live on the feast.dev website, not our docs so you can remove this change.

franciscojavierarceo
franciscojavierarceo reviewed Jan 14, 2026
View reviewed changes
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@franciscojavierarceo franciscojavierarceo franciscojavierarceo left review comments

Assignees

@franciscojavierarceo franciscojavierarceo

Copilot code review Copilot

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Add blog post on new transformation framework

2 participants

@franciscojavierarceo