Full Job Description
The Role
As an Infrastructure Engineer at Relace, you'll design and operate the systems that power our high-performance inference and training infrastructure. You'll work closely with our research and product teams to ensure our models run at scale with reliability, speed, and cost-efficiency. This is a hands-on engineering role where you'll shape how we build and scale the backbone of modern code generation.
You'll have the opportunity to:
- Architect and manage the infrastructure powering our ultra-fast inference and training stack.
- Build reliable, efficient systems for deploying and scaling ML workloads globally.
- Work on GPU scheduling, distributed systems, and high-performance cloud deployments.
- Optimize performance and cost across compute, networking, and storage layers.
- Collaborate with world-class engineers to push the limits of what small models can do.
Requirements
2+ years of experience writing high-quality production code
Strong experience with cloud infrastructure (AWS, GCP, Azure, or equivalent)
Experience with data science and systems optimization
Familiarity with ML infrastructure, GPU's, etc. a plus
Work out of our SF office in FiDi
