RL Environment Infrastructure Engineer — Sandbox & Scale

Commergence; Colorado; Hybrid; USD 150,000 - 210,000; Full time

Job summary

Commergence is seeking an AI Lab Tech Engineer to own the execution and sandboxing layer for RL environments. You will design systems to snapshot and restore environment state, enabling paused, resumed, and branched runs at training scale.

You will work with research teams and customers to deploy reliable infrastructure, optimize performance, and extend environments across multi-node, long-horizon tasks. Hybrid onsite involvement in the Bay Area is expected.

Qualifications

• Experience building production systems or research infrastructure at scale.
• Deep comfort with the systems layer: containers and isolation, filesystems, state management.
• Experience optimizing performance, cost, and scalability.
• Proficiency with cloud platforms (GCP, AWS) and distributed computing.
• Strong engineering fundamentals and reliability practices.

Responsibilities

• Design and own the sandboxing and execution layer for RL environments; enable snapshot, restore, inspect and branch capabilities.
• Develop failure-detection and rollback mechanisms to maintain known-good states in long-running rollouts.
• Extend execution to long-horizon, multi-node environments across tools and services.
• Own platform performance, scheduling, and cost-per-rollout at scale; drive observability.
• Build and maintain framework to specify, package, and deploy RL environments for researchers and customers.
• Collaborate with research teams and enterprise customers to productionize environment infrastructure.

Skills

Production systems

Distributed systems

Container/sandboxing

Filesystems

Cloud platforms

Python

Rust/Go/C

Communication with research teams

Job description

Commergence is seeking an AI Lab Tech Engineer to own the execution and sandboxing layer for RL environments. You will design systems to snapshot and restore environment state, enabling paused, resumed, and branched runs at training scale.

You will work with research teams and customers to deploy reliable infrastructure, optimize performance, and extend environments across multi-node, long-horizon tasks. Hybrid onsite involvement in the Bay Area is expected.