Researcher, Alignment Training

Responsibilities

• Develop synthetic data methods to teach models higher-level behavioral tendencies like intent understanding, reasoning, and honesty.
• Study how pre-training, mid-training, and post-training stages shape downstream model behavior.
• Build evaluation loops that connect model behavior back to training data and objectives to enable faster iteration.
• Design reusable data generation and filtering pipelines to improve training data quality and robustness.
• Create experiments to distinguish durable learned behavior from benchmark gains or distribution-specific effects.
• Collaborate across pre-training, post-training, alignment, and product teams to translate research into better model behavior.

Requirements

• Strong record of technically excellent work in large-scale ML.
• Expertise in pre-training, post-training, synthetic data, model evaluation, or training infrastructure.
• Ability to move between research hypothesis formulation and engineering execution.
• Experience designing experiments where signals may be subtle, noisy, or indirect.
• Excellent communication skills across research, engineering, and product contexts.

Preferred Qualifications

• Deep interest in alignment problems and making models more useful, honest, and steerable.
• Proven ability to manage ambiguous research questions and turn them into concrete experimental programs.

Benefits

• Competitive salary ($250K – $445K) and generous equity.
• Comprehensive medical, dental, and vision insurance.
• 401(k) retirement plan with employer match.
• Paid parental leave and flexible PTO.
• Daily meals in the office and meal delivery credits.
• Annual learning and development stipend.
• Relocation support for eligible employees.

About the Company

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of AI capabilities and seek to safely deploy them to the world through our products.