RL Environments Architect
Remote • United States - Remote • FullTime
Posted 2mo ago
About the job
Surge is building a platform to power the next generation of AI, aiming to achieve AGI by combining human intelligence and creativity with cutting-edge tools. As an RL Environments Architect, you will be responsible for designing, instrumenting, and governing the simulated worlds where AI agents learn. This involves defining the core components, reward structures, interfaces, and telemetry that ensure faithful, stable, and scalable training signals. You will also establish standards for data quality and reproducibility across large-scale agent environments, focusing on simulation fidelity, coverage, and failure surfaces to create learnable curricula that drive safe and rapid progress in agentic systems.
Responsibilities
- Architect a modular environment framework with clear APIs, curriculum scaffolds, and configurable reward/termination schemas.
- Establish quality bars for environment outputs and agent experience buffers, including coverage metrics, invariance checks, and trace audits.
- Instrument rich telemetry for episode rollouts to mitigate reward hacking, mode collapse, and exploitable loopholes.
- Partner with researchers to translate real-world tasks into robust simulations, including synthetic data generators and evaluation suites.
Requirements
- Experience building RL environments or simulators (e.g., custom physics, multi-agent, tool APIs) with an eye for determinism, performance, and observability.
- Strong instincts for designing reward functions, scenario taxonomies, and QA pipelines that keep signals aligned and drift-free.
- Comfort collaborating across research and engineering to ship pragmatic, testable environments that evolve with model capabilities.