Member of Technical Staff Post Training Reinforcement Learning and Generalization
Toronto-founded enterprise AI company building foundation models, agentic AI, enterprise search, and private deployment products.
About Cohere
Cohere provides enterprise-focused AI models and end-to-end products, including North for agentic workplace AI, Compass for enterprise search, and models for generation and retrieval. It supports private, VPC, and on-premises deployment options.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will design and run large-scale reinforcement-learning post-training experiments for language models. You will develop methods for generalization, create held-out evaluations, diagnose training failures, and convert successful research into robust and reproducible production training recipes.
Requirements
- Research
- Software engineering
- LLM training
- Reinforcement learning
- PPO
- GRPO
- RLHF
- RLVR
- Generalization
- Robustness
- Distribution shift
- Experimental design
- Distributed training
- JAX
- PyTorch
- FSDP
- Megatron
- Data quality
- Communication
Responsibilities
- Design, run, and analyze large-scale reinforcement-learning post-training experiments
- Develop methods for out-of-distribution generalization
- Build held-out evaluations for unseen tasks, domains, and formats
- Diagnose training dynamics, reward over-optimization, instability, and capability regressions
- Turn successful research into reproducible production training recipes
Benefits
- Weekly lunch stipend of $75/£75 or local equivalent
- Health and dental benefits, including a mental-health budget
- RRSP matching, 401K, and Pension Scheme
- 100% parental-leave top-up for up to 6 months
- Arts and culture, fitness/wellness, quality-time, and workspace-improvement benefits
- Education and learning stipend
- 6 weeks of paid vacation
- Travel budget for other offices and annual company offsite
- Co-working benefit
- $500 home-office stipend
- Daily lunch, snacks, and community events for office-based employees