Repository navigation
Update feast init demo #5478
Description
Is your feature request related to a problem? Please describe.
The Feast Init should create 3 projects for demonstration:
- A predictive / Tabular one (driver risk as it exists today)
- This should be extended to include ODFV transform on write and Python / Pandas Transformations
- A RAG project (exists today)Describe the solution you'd like
- A recommendation engine using vector similarity search
Describe alternatives you've considered
N/A
Additional context
N/A
Reacted by Vanshika Vanshika
Activity
franciscojavierarceo commentedon Jun 30, 2025
FYI @ntkathole
aveerapareddy commentedon Mar 7, 2026
Hi,
I reviewed PR #5946 and noticed there is still active discussion around the default
feast initbehavior, dependency impact for the RAG template, and operator compatibility.Since the scope here seems to still be evolving, I just wanted to check whether there is a specific follow-up task from this issue that would be a good place for a new contributor to help.
I’d be happy to contribute once the next concrete piece is identified.
feat(cli): Add recommendation template for feast init (feast-dev#5478)
feat(cli): Add recommendation template for feast init (feast-dev#5478)
feat(cli): Add recommendation template for feast init (feast-dev#5478)
mitapujari-05 commentedon Sep 24, 2026
hello , i would like to work on this issue. I'll start by adding the recommendation engine demo using vector similarity search. Will that be okay? as a first step?
docs: Describe the on demand feature views added to the local template
feat: Add native Python mode on demand feature view to the local temp…
fix: Create sqlite tables for on demand feature views with write_to_o…
feat: Add transform-on-write on demand feature view to the local temp…
test: Verify transform-on-write features are materialized by the loca…
docs: Describe the on demand feature views added to the local template
throw29 commentedon Oct 8, 2026
Hi! I’d like to contribute to this issue. The current local template already includes a Pandas
OnDemandFeatureViewexample. Could you clarify the smallest remaining slice for “transform on write” and the expected user-facing workflow or acceptance criteria?