Senior Software Engineer, ML Infrastructure

We're looking for a Senior Software Engineer to join our ML Infrastructure team and support the foundational infrastructure that powers Gridmatic.

Our platform challenges are shaped by the nature of energy markets: forecasts and trading decisions run on tight schedules, battery dispatch commands must execute reliably in real time, and ML models need to train and deploy continuously as new data arrives.

• Design and build the compute platform that runs Gridmatic's production services, batch jobs, and ML training workloads
• Help own our production workflow orchestration end to end, from the cluster and node pools it runs on to the abstractions our teams build on top of it
• Make our ML iteration cycle faster and cheaper by profiling workflows, cutting latency, and improving how we use compute
• Build the observability and developer tooling that lets engineers understand and troubleshoot their own workloads, from post-run cost reporting to monitoring and alerting
• Improve the reliability and resiliency of production workloads, including how we handle capacity constraints and multi-region routing
• Manage GPU and accelerator capacity: node pools, drivers, spot vs. on-demand tradeoffs, and scheduling and quota so training and batch jobs get the compute they need without overspending
• Own autoscaling and quota management so the platform scales up under load and down to zero when idle
• Work closely with the ML team to find pain points and quickly ship solutions
• Make architectural decisions that shape how we build software as we grow

• Significant experience building and operating production infrastructure on a public cloud platform (we run on GCP, but AWS or Azure experience translates well)
• Strong distributed systems and infrastructure skills: standing up services, scaling and debugging Kubernetes (we run on GKE and it's foundational to our platform), writing Terraform, and comfort with cloud networking, IAM, and secrets management