Distributed Systems · GPU
WHAT YOU'LL DO
As a Principal Software Engineer, you will help set the technical direction for a core part of Radiant's AI cloud, joining one of four teams: Infra & Cluster, Observability, MLOps Training, or MLOps Inference. We'll match you to the team that fits your experience.
You'll lead the design and delivery of the distributed systems that run GPU infrastructure and AI workloads at scale, partnering closely with product, infrastructure, SRE, networking, and storage teams, and with customers running the most demanding workloads.
This is a hands-on technical leadership role, delivered through architecture, code, and the standards other teams build on.
The teams
- Infra & Cluster: provisioning, lifecycle, and cluster management behind our Bare Metal and Virtual Machine products
- Observability: FlightDeck, our platform for observing and operating AI infrastructure, and the telemetry behind it
- MLOps Training: the platform customers use to train models on our GPU clusters, including Managed Kubernetes and Managed Slurm
- MLOps Inference: the platform customers use to serve models in production
KEY RESPONSIBILITIES
- Set the technical vision and architecture for your team's area, and lead its design and implementation.
- Design and build distributed systems that run reliably on GPU infrastructure at scale.
- Translate product and customer requirements into robust architectures, APIs, and technical designs, working closely with Product.
- Own the performance, reliability, and cost of your team's systems as the platform grows.
- Partner with SRE,