Staff DevOps Engineer

Nexxa.AI•Sunnyvale, CA

About The Position

Nexxa is seeking a Senior/Staff DevOps Engineer to build and operate the infrastructure for AI and industrial systems. This role involves ensuring the reliability, observability, and resilience of production ML and data workloads, from GPU-backed clusters to pipelines connecting to industrial environments. The ideal candidate will have deep infrastructure ownership experience, where uptime and reliability directly impact physical operations. You will collaborate with AI, data, and product engineering teams to ensure their systems run safely and at scale in production.

Requirements

6+ years of experience in DevOps, Site Reliability Engineering, Platform Engineering, or infrastructure-focused software engineering roles
Deep hands-on experience with Cloud platforms (AWS, GCP, or Azure) at production scale
Kubernetes in production, including GPU workload scheduling
Infrastructure-as-code tooling (Terraform, Pulumi, or equivalent)
CI/CD systems (e.g., GitHub Actions, GitLab CI, CircleCI, Jenkins, ArgoCD)
Strong track record designing and operating observability stacks (e.g., Prometheus, Grafana, Datadog, OpenTelemetry)
Excellent scripting/programming skills (Python, Go, or Bash) for automation and tooling
Proven ability to independently scope and lead infrastructure projects from design through production rollout
Strong incident management instincts — you can lead through an outage calmly and drive toward root cause

Nice To Haves

Experience supporting ML/AI infrastructure — training clusters, model serving, data pipelines
Experience operating infrastructure that bridges cloud and edge/on-prem environments, especially in industrial or manufacturing contexts
Familiarity with data warehouse/lakehouse platforms (Snowflake, BigQuery, Redshift, Databricks)
Experience with service mesh, zero-trust networking, or compliance frameworks relevant to industrial/critical infrastructure (e.g., SOC 2, IEC 62443)
History of building internal developer platforms or self-service infrastructure tooling
Experience scaling infrastructure teams or setting technical direction at a Staff level

Responsibilities

Own and evolve Nexxa's core infrastructure — compute, networking, storage, and deployment systems — end-to-end
Design and operate CI/CD pipelines that support fast, safe iteration across AI, data, and product engineering teams
Build and maintain infrastructure-as-code (e.g., Terraform, Pulumi) for reproducible, auditable environments across cloud and on-prem/edge deployments
Architect and manage Kubernetes-based platforms for training, inference, and application workloads, including GPU scheduling and autoscaling
Partner with data and AI teams to support the infrastructure behind: Data warehouses and lakehouse architectures (e.g., Snowflake, BigQuery, Redshift, Databricks), Feature stores, embedding indices, and retrieval pipelines, Model training, evaluation, and serving infrastructure
Define and drive observability practices — metrics, logging, tracing, and alerting — across distributed systems
Establish and enforce reliability practices: SLOs/SLIs, incident response, postmortems, and on-call rotations
Design for security and compliance across cloud infrastructure, secrets management, and access control, particularly relevant to industrial and legacy-environment integrations
Make pragmatic tradeoffs across cost, latency, reliability, and developer velocity
Collaborate with engineering leadership to define infrastructure roadmap and platform strategy
Mentor engineers on infrastructure best practices and raise the bar for operational excellence across the org