Senior AI Inference & Runtime Platform Engineer

AZX; Seattle, Northern (WA, KY); Hybrid; USD 150,000 - 190,000; Full time

Benefits offered by this job

Health insurance

Equity

Fully remote culture with Seattle team

Flexible paid time off

Bonus eligibility

Job summary

AZX is pioneering AI-driven transformations across climate and sustainability. We are seeking an experienced engineer to own the machine-layer stack, including inference control planes and a secure sandboxing platform.

You will write Rust, build Kubernetes controllers, and craft FastAPI endpoints while scaling across managed GPU clouds and customer-managed clusters. The role emphasizes safety, least privilege, and auditable actions within a distributed system stack.

Qualifications

• 5+ years shipping production systems in a systems language (Rust preferred).
• Experience with Kubernetes controllers/operators and autoscaling.
• Ability to threat-model isolation boundaries and enforce least privilege in sandboxes.
• Hands-on with open-weight LLM serving infra and production endpoints.
• Proven performance discipline in distributed systems.

Responsibilities

• Manage the serving tier for open-weight models, including deployment, config, and SLOs.
• Administer Kubernetes layer for inference and sandbox workloads, autoscaling, and GPU scheduling.
• Own end-to-end stateful data plane with reliable restore procedures.
• Oversee sandbox runtime, isolation boundary, and control plane activities.
• Direct FastAPI control-plane services and IaC tooling (Terraform/OpenTofu, Bicep).
• Support backend services with observability and reproducible builds.
• Maintain OSS posture with proper documentation and release processes.

Skills

Rust

Kubernetes

Threat modelling

LLM serving

Distributed systems

Cloud infra

Terraform/OpenTofu

GPU tooling

Education

Bachelor's Degree

Master's (nice to have)

Tools

KEDA

Karpenter

Kubebuilder/CRDs

OpenTelemetry

Firecracker/Kata

Qdrant/pgvector

Job description

AZX is pioneering AI-driven transformations across climate and sustainability. We are seeking an experienced engineer to own the machine-layer stack, including inference control planes and a secure sandboxing platform.

You will write Rust, build Kubernetes controllers, and craft FastAPI endpoints while scaling across managed GPU clouds and customer-managed clusters. The role emphasizes safety, least privilege, and auditable actions within a distributed system stack.