feat: Add blog post on Feast dbt integration #5915
Conversation
Co-authored-by: franciscojavierarceo <[email protected]>
|
|
||
| # Streamlining ML Feature Engineering with Feast and dbt | ||
|
|
||
| If you're building machine learning models in production, you've likely faced the challenge of managing features consistently across training and serving environments. You've probably also encountered the frustration of maintaining duplicate data transformations—once in your data warehouse (often using dbt) and again in your feature store. |
There was a problem hiding this comment.
@copilot We should reframe this from the lens of a user of DBT that loves their models and wants to easily get some dbt model into production.
There was a problem hiding this comment.
Reframed the entire blog post from a dbt user's perspective. The post now:
- Opens by speaking directly to dbt users who've invested in their models
- Emphasizes that existing dbt models are already ML-ready and don't need rewriting
- Frames steps as using "existing" models rather than creating new ones
- Adds a "What Just Happened?" section to highlight the value of going from dbt to production
- Updates the conclusion to be action-oriented for dbt users
The narrative now centers on dbt users who love their models and want to easily get them into production for ML without duplication. ✨
Commit: 48768fd
Co-authored-by: franciscojavierarceo <[email protected]>
Co-authored-by: franciscojavierarceo <[email protected]>
|
Just as a heads up, I was blocked by some firewall rules while working on your feedback. Expand below for details. Warning Firewall rules blocked me from connecting to one or more addresses (expand for details)I tried to connect to the following addresses, but was blocked by firewall rules:
If you need me to access, download, or install something from one of these locations, you can either:
|
franciscojavierarceo
left a comment
There was a problem hiding this comment.
@copilot please replace instances of ML/machine learning with AI.
Co-authored-by: franciscojavierarceo <[email protected]>
Replaced all instances of ML/machine learning with AI throughout the blog post. Changes include:
The blog post now consistently uses AI terminology while maintaining all technical content and the dbt-user-centric perspective. Commit: 564a01f Warning Firewall rules blocked me from connecting to one or more addresses (expand for details)I tried to connect to the following addresses, but was blocked by firewall rules:
If you need me to access, download, or install something from one of these locations, you can either:
|
What this PR does / why we need it:
Adds blog post documenting Feast's dbt integration to
infra/website/docs/blog/feast-dbt-integration.md, written from the perspective of dbt users who want to leverage their existing models for production AI.Content structure:
manifest.jsonNarrative approach:
The blog post speaks directly to dbt users, emphasizing that their existing models are already AI-ready and can go to production without rewriting. All examples frame steps as using "existing" models rather than creating new ones, positioning Feast as respecting dbt as the single source of truth. Uses AI terminology consistently throughout (e.g., "AI features", "AI team", "AI predictions", "AI deployment").
Misc
Blog post validated with Astro build system. Uses existing hero image asset. Authored by Francisco Javier Arceo and Yassin Nouh.
Original prompt
💡 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.