Staff Engineer, Post-Training RL for LLMs

Cohere; Toronto, Montreal; Hybrid; GBP 90,000 - 150,000; Full time

Benefits offered by this job

Lunch stipend

Health benefits

Mental health budget

RRSP matching

Pension plan

Parental leave top‑up

Enrichment benefits

Vacation 6 weeks

Travel budget

Offsite

Job summary

Cohere is seeking a Member of Technical Staff in Post-Training to advance RL-based training for large language models. You will design training recipes, environments, and evaluations, collaborating with data infrastructure and pre-training teams. A strong focus on generalization and robust training pipelines drives your day-to-day work.

The role offers remote-friendly work with offices in major cities and a supportive benefits package for researchers and engineers.

Qualifications

• Experience training LLMs with RL methods and large-scale experiments.
• Strong publication record or shipped models showing generalization.
• Rigorous experimental practice and clear conclusions from noisy results.
• Experience with distributed training and modern ML tooling.
• Excellent written and verbal communication skills.

Responsibilities

• Design, run, and analyse large-scale RL post-training experiments.
• Develop methods to improve generalization and robustness.
• Build held-out evaluations for unseen tasks and domains.
• Diagnose training dynamics and model failure modes.
• Turn research into robust, reproducible training recipes.

Skills

LLM training

Reinforcement learning

Research experience

Communication

Data quality

Education

PhD/masters in ML or CS

Tools

JAX

PyTorch

Megatron

FSDP

Job description

Cohere is seeking a Member of Technical Staff in Post-Training to advance RL-based training for large language models. You will design training recipes, environments, and evaluations, collaborating with data infrastructure and pre-training teams. A strong focus on generalization and robust training pipelines drives your day-to-day work.

The role offers remote-friendly work with offices in major cities and a supportive benefits package for researchers and engineers.