DevOps / Site Reliability Engineer

Bespoke Labs

23h•Remote

About The Position

Bespoke Labs is an AI research and data company building the datasets, benchmarks, and evaluation infrastructure that power frontier AI models. We're backed by leading investors, trusted by top AI labs, and have research accepted at venues like ICLR 2026. Our team is small, moves fast, and has an outsized impact on how the next generation of AI is built. We're looking for a mid-level DevOps / Site Reliability Engineer to own and scale our cloud infrastructure. You'll work closely with engineering and ML teams to keep our systems reliable, observable, and fast — directly supporting the infrastructure that powers AI data pipelines at scale.

Requirements

3–5 years in DevOps, SRE, or infrastructure engineering
Strong AWS experience — EKS, EC2, RDS, S3, IAM
Kubernetes — deployment, scaling, troubleshooting in production
CI/CD pipelines — GitHub Actions, ArgoCD, or similar
Infrastructure as Code — Terraform, Pulumi, or CDK
Python or Go scripting
Experience working in production environments with real users
Comfort with ambiguity and ability to operate autonomously

Nice To Haves

Experience supporting ML training workloads or GPU clusters
Familiarity with distributed computing or large-scale data pipelines
Prior work at an AI, ML, or data company
Open-source contributions or published technical writing

Responsibilities

Own cloud infrastructure on AWS — EC2, EKS, RDS, S3, IAM, VPC
Manage Kubernetes clusters and container orchestration end-to-end
Build and maintain CI/CD pipelines using GitHub Actions or similar
Implement monitoring, alerting, and observability stacks (Prometheus, Grafana, or DataDog)
Improve reliability, performance, and security of production systems
Automate infrastructure with Terraform or similar IaC tools
Debug and resolve issues across complex, distributed systems
Participate in design reviews and help raise the infrastructure bar