Staff Software Engineer, ML Feature Store

Snap Inc is a technology company. We believe the camera presents the greatest opportunity to improve the way people live and communicate. Snap contributes to human progress by empowering people to express themselves, live in the moment, learn about the world, and have fun together.

The Company operates Snapchat, a visual messaging app that enhances your relationships with friends, family, and the world, and Specs Inc., a wholly-owned subsidiary dedicated to making computing more human, in addition to Bitmoji, Saturn, and other digital services.

We’re looking for a Staff Software Engineer to join Snap Inc on our Feature Store team!

This tech lead will be responsible for the design of the team's systems and data pipelines, establishing clear abstractions and reusable components so the team builds on common foundations rather than duplicating work. You'll drive development of the core services and data flows — including how data is stored, retrieved and served for both high-throughput and low-latency use cases — while ensuring they meet correctness, performance, and reliability requirements. A central part of the role is enforcing engineering quality through validation, testing, monitoring, and observability and owning the operational health of these systems from cost efficiency to on-call and incident response. Throughout, you'll set technical direction, review designs and code, and mentor engineers, raising the bar on craftsmanship across the team.

Serve features reliably for large-scale batch inference and low-latency online inference, maintaining consistency across serving paths and meeting latency and freshness requirements of production models.

Maintain features in a centralized, versioned feature registry, ensuring definitions are discoverable, well-documented and reusable across teams and models so common logic is defined once rather than duplicated.

Design and manage feature storage solutions for correctness, freshness and efficient retrieval, selecting appropriate storage patterns for training-scale reads and low-latency serving lookups.

Build scalable, distributed systems with attention to compute and storage efficiency, backed by monitoring, alerting, and sound operational practices that keep feature serving systems reliable and maintainable in production.