MLOps Engineer

Zof AI; San Francisco, Northern (CA, KY); Hybrid; USD 170,000 - 240,000; Full time

Job summary

Zof AI seeks an experienced MLOps Engineer to own the platform for AI systems, spanning model serving, scaling, and reliability. You will push the platform that enables a small team to run large-scale AI workloads efficiently in production, with a focus on GPU throughput, observability, and cost control.

As a senior on-site engineer in San Francisco, you will design and operate the backend infrastructure, partner with software engineers, and drive incident practices, rollback automation, and

Qualifications

• Experience operating production AI/ML systems.
• Strong infra and systems engineering fundamentals.
• Experience with cloud platforms, containers, and orchestration.
• Experience with observability and reliability tooling.
• Judgment about cost, performance, and operational trade-offs.
• Clear written and verbal communication.
• High ownership of production systems.

Responsibilities

• Design and operate model serving, scaling, and compute infrastructure.
• Own GPU and inference efficiency, capacity, and cost.
• Build the platform for provisioning, running, and scaling agents.
• Build observability for AI systems: tracing, monitoring, and alerting.
• Keep production AI systems stable, debuggable, and recoverable.
• Automate deployment, rollback, and environment management.
• Set reliability standards and incident practices for AI workloads.
• Partner with engineers to make the platform fast to build on.

Skills

Production AI

Cloud platforms

Containers

Observability

Reliability tooling

Ownership

Clear communication

Tools

Kubernetes

Terraform

Docker

Job description

Zof AI is seeking a MLOps Engineer to own the platform our AI systems run on. This is a consolidated platform role spanning what the market posts as AI Infrastructure, MLOps, LLMOps, and agent platform engineering: model serving and scaling, GPU and compute efficiency, and the provisioning, observability, and reliability infrastructure that keeps production agents stable. The ideal candidate has operated real AI workloads in production and builds infrastructure that lets a small team run systems well above its weight.

Engineering · Senior · Full-time · On-site · San Francisco, CA

Responsibilities

• Design and operate model serving, scaling, and compute infrastructure.
• Own GPU and inference efficiency, capacity, and cost.
• Build the platform for provisioning, running, and scaling agents.
• Build observability for AI systems: tracing, monitoring, and alerting.
• Keep production AI systems stable, debuggable, and recoverable.
• Automate deployment, rollback, and environment management.
• Set reliability standards and incident practices for AI workloads.
• Partner with engineers to make the platform fast to build on.

Requirements

• Experience operating production AI, ML, or high-scale backend systems.
• Strong infrastructure and systems engineering foundation.
• Experience with cloud platforms, containers, and orchestration.
• Experience with observability and reliability tooling.
• Judgment about cost, performance, and operational trade-offs.
• Clear written and verbal communication.
• Comfort operating in a fast-moving environment.
• High ownership of systems in production.

Nice to have

• Experience with GPU clusters, inference servers, or model gateways.
• Experience running agent workloads or long-lived AI processes.
• Experience with Kubernetes, Terraform, or similar tooling.
• Experience in early-stage platform teams.

Experience operating production AI or ML systems at scale is required