Remote Staff Engineer, AI RL Environments

Cohere; Greater London; Remote; GBP 135,000 - 300,000; Full time

Benefits offered by this job

Weekly lunch stipend

Health, dental, and mental health

RRSP/401K/Pension

Parental leave top-up

Education & learning stipend

Paid vacation (6 weeks)

Office stipend for travel to other off

Home office stipend

Job summary

Cohere is seeking a Staff-level ML/RL Environments engineer to design and run reinforcement learning environments for enterprise tasks. You will train and evaluate agents, improve environments, and collaborate with product teams to close performance gaps.

The role emphasizes building scalable RL workflows, automating evaluation, and working with external vendors to ensure high data and verifier quality. Remote-friendly with global offices.

Qualifications

• Engineered agents for industry use cases.
• Reviewed agent trajectories to pinpoint failure points.
• Measured agentic capabilities and turned insights into repeatable processes.
• Translated desired outcomes into verifier implementations and reward schemes.
• Designed and ran annotation workflows to surface insights and data quality.
• Built synthetic data pipelines to scale evaluation and training.
• Used agents in daily work and improved setups for higher productivity.
• Familiarity with RL: scaling, troubleshooting and tuning environments (a plus).

Responsibilities

• Build new RL environments for different agentic capabilities and industries.
• Train and evaluate agents inside these environments.
• Integrate tasks, data, tool implementations, and verifiers.
• Identify gaps in agent performance and improve agents and environments.
• Collaborate with external vendors to create high-quality RL environments and tooling.
• Automate discovery of model capability gaps and measure agent performance during evals and training.

Skills

Agent development

RL environment design

Agent evaluation

Reward design

Data annotation workflows

Synthetic data pipelines

Model training

Experimentation

Job description

Cohere is seeking a Staff-level ML/RL Environments engineer to design and run reinforcement learning environments for enterprise tasks. You will train and evaluate agents, improve environments, and collaborate with product teams to close performance gaps.

The role emphasizes building scalable RL workflows, automating evaluation, and working with external vendors to ensure high data and verifier quality. Remote-friendly with global offices.