Senior Reliability Engineer – AI Infrastructure

Fireworks AI; United States; On-site; USD 140,000 - 210,000; Full time

Job summary

Fireworks AI seeks a Reliability Engineer to own and drive production resiliency across GPU-accelerated inference and training systems. You will define SLOs, error budgets, and readiness criteria while leading incident response and postmortems.

Collaboration with cloud infra, ML infrastructure, and product teams is essential to ensure scalable, fail-fast systems. You will automate observability, implement standard tooling, and reduce toil with robust reliability practices across multi-region

Qualifications

• 5+ years with Linux internals, system performance troubleshooting, and networking fundamentals.
• Bachelor’s or Master’s in CS/CE or equivalent practical experience.
• Experience with GPU and ML infrastructure exposure is a plus.
• Strong emphasis on incident response and postmortems.

Responsibilities

• Define reliability standards: SLOs, error budgets, production readiness criteria.
• Own incident management, postmortems, and observability standards across services.
• Automate failure testing, alerting, and runbooks to reduce toil.
• Collaborate with cloud infra, AI systems, and product teams to ensure resiliency.

Skills

Linux internals

System performance troubleshooting

Networking fundamentals

Python

Go

C++

Rust

Kubernetes

Terraform

Docker

Prometheus

Grafana

OpenTelemetry

SRE practices

Incident response

Education

Bachelor's or Master's in Computer Science, Computer Engineering, or equivalent

Tools

Kubernetes

Terraform

Docker

Prometheus

Grafana

OpenTelemetry

Job description

Fireworks AI seeks a Reliability Engineer to own and drive production resiliency across GPU-accelerated inference and training systems. You will define SLOs, error budgets, and readiness criteria while leading incident response and postmortems.

Collaboration with cloud infra, ML infrastructure, and product teams is essential to ensure scalable, fail-fast systems. You will automate observability, implement standard tooling, and reduce toil with robust reliability practices across multi-region