RL Environment Engineer for Conversational AI Training

BeSimple AI; United States; On-site; USD 120,000 - 190,000; Full time

Job summary

Besimple AI is building datasets, benchmarks, and training environments to help voice models understand people and interact naturally. We seek an engineer with hands-on experience in building RL environments to turn real-world interactions into taskable challenges for agents.

You will define tasks, observations, actions, and rewards, and design environments that support scalable training pipelines and verifiable outcomes.

Qualifications

• Hands-on experience building or extending RL environments.
• Strong engineering skills and tested, maintainable software.
• Experience defining task interfaces, managing environment state, implementing rewards, and debugging agent-environment interactions.
• Ability to investigate why an agent fails and improve the environment.
• Clear technical communication and ability to take prototypes to reliable systems.
• Experience with speech/audio or full-duplex interaction is a plus.
• Familiarity with tool-use, coding, or browser/computer-use environments.

Responsibilities

• Build training environments.
• Define tasks, observations, actions, state transitions, rewards, and termination conditions.
• Create environments where agents ask questions, use tools, and reach verifiable outcomes.
• Design rewards and verifiers.
• Implement checks for task completion and progress.
• Turn human interactions into tasks and construct scenarios from demonstrations.
• Build interactive simulators and user/tool simulators that scale.
• Develop rollout interfaces, parallel execution, logging, replay, and debugging tools.

Skills

RL environments

Engineering skills

Software testing

Debugging RL

Task interfaces

Environment state

Rewards design

Technical communication

Speech/audio experience

Tools

Gymnasium

RL frameworks

Simulation tools

Job description

Besimple AI is building datasets, benchmarks, and training environments to help voice models understand people and interact naturally. We seek an engineer with hands-on experience in building RL environments to turn real-world interactions into taskable challenges for agents.

You will define tasks, observations, actions, and rewards, and design environments that support scalable training pipelines and verifiable outcomes.