LLM Alignment & Post-Training Research Scientist

Scale AI; New York (NY); On-site; USD 150,000 - 200,000; Full time

Benefits offered by this job

Health & wellbeing benefits

Learning & development stipend

Community & ERG events

Parental support

Job summary

Scale AI seeks Research Scientists and Research Engineers to advance LLM post-training techniques (SFT, RLHF, reward modeling). The role emphasizes optimizing data curation and algorithms to improve instruction following, factual accuracy, coding, multilingual and multimodal understanding.

You will develop novel methods, collaborate with researchers, and publish in top conferences. Join Scale AI to help shape the next generation of generative models and work with leading labs on strategic inputs

Qualifications

• Published research in ML at major conferences or journals.
• Deep understanding of DL, RL, and large-scale model fine-tuning.
• Excellent written and verbal communication skills.
• Ph.D. or Master’s in CS/ML/AI or related field.
• Experience in customer-facing roles.
• Experience with post-training techniques such as RLHF or instruction tuning.

Responsibilities

• Develop novel post-training methods to improve LLMs.
• Collaborate with researchers and engineers to set data-driven best practices.
• Optimize data curation and algorithms to boost instruction following, factuality, and multilingual/multimodal understanding.
• Publish findings at top AI conferences and contribute to model robustness.

Skills

Post-training techniques

Reinforcement learning

Deep learning

Communication skills

Customer-facing experience

Education

Ph.D. or Master’s in CS/ML/AI

Job description

Scale AI seeks Research Scientists and Research Engineers to advance LLM post-training techniques (SFT, RLHF, reward modeling). The role emphasizes optimizing data curation and algorithms to improve instruction following, factual accuracy, coding, multilingual and multimodal understanding.

You will develop novel methods, collaborate with researchers, and publish in top conferences. Join Scale AI to help shape the next generation of generative models and work with leading labs on strategic inputs