Bespoke Labs
RL Environments Engineer
About this role
Bespoke Labs seeks an RL Environments Engineer to build the infrastructure and pipelines that transform environment ideas into validated agentic coding tasks at scale. You'll own the end-to-end production of RL environments—from templating and automation to grading and quality assurance—focusing on shipping volume while maintaining rigor.
What you'll do
- Design and build environment-generation pipelines that produce RL tasks programmatically with templating, automated grading, and verification
- Create high-fidelity coding environments based on real codebases, dependencies, and production software conventions
- Scale agentic task creation from dozens to hundreds or thousands through automation and infrastructure improvements
- Identify and eliminate bottlenecks in environment production by building tools and internal systems that increase throughput
- Manage the full task lifecycle: prompt design, environment setup, grader development, model evaluation, failure analysis, and iteration
- Defend task quality at scale by detecting reward hacking, grader exploits, and maintaining verification standards
What they're looking for
- Python and strong software engineering fundamentals
- Production software experience: large codebases, build systems, testing, DevOps, and RCA
- RL environment and agentic task design and validation
- Scaled pipeline and infrastructure development
- GCP experience for large-scale workloads
- Sandboxed code execution and verification systems
- Understanding of frontier coding agent capabilities and limitations
- Tool-building and automation mindset
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Bespoke Labs
Bespoke Labs builds specialized AI agents and reinforcement learning platforms that enable enterprises and researchers to develop, train, and deploy advanced AI systems at scale. The company is hiring AI engineers, product engineers, infrastructure engineers, full stack engineers, and research engineers to develop agent capabilities, build accessible RL platforms, and create production systems for complex agentic workflows.
View all jobs at Bespoke LabsLikely interview questions
- Walk us through a time you built an environment or task-generation pipeline from scratch—what were the key scaling bottlenecks, and how did you solve them?
- Describe your experience validating RL environments or agentic tasks at scale. How did you catch reward hacking or grader loopholes?