Research Engineer, RL Environments & Training Infra

Preference Model; Seattle (WA); On-site; USD 200,000 - 350,000; Full time

Benefits offered by this job

Competitive cash and equity compensation (>90th percentile)

Ownership and autonomy

Health, vision, dental benefits

401K match

Lunch provided everyday onsite

Weekly snack orders

Visa sponsorship & relocation support

Job summary

Preference Model is seeking a Research Engineer or Research Scientist located in Seattle, WA, to innovate in the field of self-directed learning for models. The role involves training and evaluating models in custom RL settings and optimizing ML infrastructure.

Ideal candidates will have experience with LLM post-training pipelines and be proficient in Python and modern RL frameworks. Preference Model offers a competitive compensation package, the opportunity to work with top engineers, and a challenging startup environment.

Qualifications

• Experience running end-to-end LLM post-training pipelines.
• Proficiency in Python and PyTorch or JAX.
• Experience with at least one modern RL training framework.
• Experience building and operating ML infrastructure at scale.

Responsibilities

• Train and evaluate models on proprietary RL environments.
• Architect and optimize RL training infrastructure.
• Design, implement, and test training environments for RL agents.
• Profile and optimize training runs end-to-end.

Skills

End-to-end LLM post-training pipelines

Python

PyTorch or JAX

Modern RL training framework

ML infrastructure at scale

Job description

Preference Model is seeking a Research Engineer or Research Scientist located in Seattle, WA, to innovate in the field of self-directed learning for models. The role involves training and evaluating models in custom RL settings and optimizing ML infrastructure.

Ideal candidates will have experience with LLM post-training pipelines and be proficient in Python and modern RL frameworks. Preference Model offers a competitive compensation package, the opportunity to work with top engineers, and a challenging startup environment.