Model-Behavior Research Engineer: Evals & Alignment

OpenAI; San Francisco (CA); On-site; USD 310,000 - 460,000; Full time

Benefits offered by this job

Equity offers

Job summary

OpenAI is looking for a Research Engineer/Scientist specializing in model behavior in San Francisco, CA. As a researcher, you will create evaluations that steer the behavior of models, collaborating across various domains including alignment and training.

Your impact on today’s products and future AGI will be profound as you design evaluations to ensure alignment with intended behaviors. Ideal candidates will have extensive experience in machine learning and a strong intuition for product requirements. Compensation ranges from $310K to $460K with equity offers.

Qualifications

• Extensive experience navigating large ML codebases.
• Published work on language model evals from reputable labs.
• Ability to understand AGI implications.

Responsibilities

• Help decide how products and AGI behave.
• Design evaluations to align models with intended behavior.
• Develop new technology for model evaluation.

Skills

Machine Learning expertise

Evaluation design skills

Strong product intuition

Experience with language models

Artistic or design experience

Job description

OpenAI is looking for a Research Engineer/Scientist specializing in model behavior in San Francisco, CA. As a researcher, you will create evaluations that steer the behavior of models, collaborating across various domains including alignment and training.

Your impact on today’s products and future AGI will be profound as you design evaluations to ensure alignment with intended behaviors. Ideal candidates will have extensive experience in machine learning and a strong intuition for product requirements. Compensation ranges from $310K to $460K with equity offers.