AI Feature Store Migration Architect

Velmox Data

Remote · $160k – $200k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

What we already have: a production feature store serving 200+ features across 15 ML models, built on a custom internal system that's become a maintenance burden. What we're building: a migration path to a modern, open-source feature store architecture without disrupting the ML models that depend on those features in real time. As the AI Feature Store Migration Architect, you'll own the technical design and phased migration of Velmox's feature store to a Feast or Tecton-based architecture. That means designing the migration plan, building compatibility shims that let existing models continue to serve during cutover, validating feature value parity across old and new systems, and decommissioning the legacy system without downtime. This is a senior role with end-to-end ownership of a high-stakes infrastructure migration. You'll work with ML engineers who depend on features in production, data engineers who own the pipelines feeding the store, and platform leadership who set the timeline. If you've migrated production ML infrastructure before, you'll know exactly what this project demands.

Fully remote.

Responsibilities

Own the technical design and phased migration plan for the feature store
Build compatibility shims ensuring existing ML models continue serving during cutover
Develop and run feature value parity validation across old and new feature store systems
Coordinate migration timeline with ML engineering, data engineering, and platform leadership
Execute decommissioning of the legacy feature store with zero production downtime

Requirements

5+ years ML infrastructure or data platform engineering
Deep experience with feature store systems (Feast, Tecton, Hopsworks, or similar)
Strong Python, SQL, and Apache Spark skills
Track record managing infrastructure migrations without production incidents
Experience with MLflow for experiment tracking and model lineage