Description
Relay is fundamentally reshaping how goods move in an online era. As a Senior Machine Learning Engineer at Relay, you'll own a critical part of ML: productionising of our models end-to-end, from training pipeline through live integration to measured business impact.
You'll build and mature our ML Platform, evolve the model serving architecture, expand reusable components adoption, and set standards for how Relay ships ML org-wide. You'll also strengthen existing models by architecting integration and system testing within training pipelines, automated releases, and drift monitoring.
The ideal candidate has at least two years of experience deploying and operating models in production and four years of building software on high-performing teams. They're comfortable diving into unfamiliar codebases and languages to ship a model into a live system.
Our team is intellectually vibrant, with ~160 people, around half in engineering, product, and data. We have 45+ advanced degrees across computer science, mathematics, and operations research. Every parcel Relay handles is touched by ML, with 10+ models running in the critical path of a live logistics network.
Our ML Stack highlights include:
- Python and Rust
- Rust with ONNX in-process model execution
- Chalk.ai as our Feature Store
- GCP Agent Platform Endpoints for model serving
- Cloud-native on GCP, with services running on Kubernetes and extensive use of BigQuery
Benefits and perks are not mentioned in this job posting.