Platform Engineer, Multi-Tenant AI Deployments (Remote)

Runvo; United States; Remote; USD 119,000 - 185,000; Full time

Benefits offered by this job

Remote work during UK/EU hours

Job summary

Runvo is seeking a Platform Engineer to own the deployment and serving platform for private AI workspaces used by professional-services firms. Each customer runs in a dedicated environment with separate infrastructure and data stores, so the role focuses on repeatable deployments, demonstrable tenant isolation, responsive model access, and safe ongoing upgrades.

You will own provisioning, ensure isolation between customers, manage latency and costs, and roll out upgrades with minimal downtime

Qualifications

• Substantial experience operating latency-sensitive production systems relied on by others.
• Practical ownership of infrastructure as code, secrets, and identity systems.
• Experience running a shared technology stack for multiple tenants and understanding its failure modes.
• A measurement-led approach, including benchmarking to inform engineering decisions.

Responsibilities

• Build and operate the end-to-end process for provisioning single-tenant deployments.
• Maintain and verify separation between customer instances, data stores, keys, and logs.
• Manage latency and cost in the service path connecting workspaces to external model providers.
• Roll out upgrades across deployments while avoiding unplanned customer downtime.
• Instrument the platform so performance changes and regressions can be measured.

Skills

Infrastructure as code

Secrets management

Multi-tenant architecture

Performance benchmarking

Job description

Runvo is seeking a Platform Engineer to own the deployment and serving platform for private AI workspaces used by professional-services firms. Each customer runs in a dedicated environment with separate infrastructure and data stores, so the role focuses on repeatable deployments, demonstrable tenant isolation, responsive model access, and safe ongoing upgrades.

You will own provisioning, ensure isolation between customers, manage latency and costs, and roll out upgrades with minimal downtime