Software Engineer, Observability (Supercompute)
Core
Build software systems and platforms to automatically predict, monitor, and rapidly respond to operational conditions and incidents for large-scale supercompute centers running frontier AI training and inference.
Role type
Senior IC observability software engineer (supercompute infrastructure)
Builds
Production systems (UI, APIs, data pipelines), predictive alerting tools, and interactive dashboards for GPU uptime and site-wide reliability.
Domain
Supercomputing / Frontier AI Infrastructure / Datacenter Operations
Deliverable
production ML models
Required skills
Python, Rust, C++, Java, Postgres, MySQL, MongoDB, DynamoDB, state machine design, cross-functional collaboration
Preferred skills
Time-series databases (Prometheus, VictoriaMetrics, ClickHouse, InfluxDB), streaming telemetry pipelines, Grafana/New Relic/Splunk, React/Angular, Docker/Kubernetes, CI/CD (ArgoCD, GitHub Actions), HPC environments (Slurm, GPU telemetry), industrial/OT data integration (SCADA, Modbus, OPC-UA)
Responsibilities
Design and operate multi-service production systems for reliability and uptime monitoring; Develop predictive and analytical tools for supercompute center operations; Build frontend visualizations and interactive dashboards for real-time data views; Collaborate with users to incorporate feedback and improve GPU uptime.
Seniority
Senior, hands-on IC
