Senior Research Scientist, Reward Models
仕事内容
Lead cutting-edge research on reward model architectures and RLHF training for large language models.
Develop rubric-based grading and evaluation methods to improve interpretability and consistency.
Study reward hacking, specification gaming, and robust alignment strategies.
Design large-scale experiments to assess generalization, robustness, and failure modes.
Collaborate with Finetuning and Alignment Science teams to translate insights into production improvements.
Mentor researchers, publish findings, and grow institutional knowledge around reward modeling.
Develop rubric-based grading and evaluation methods to improve interpretability and consistency.
Study reward hacking, specification gaming, and robust alignment strategies.
Design large-scale experiments to assess generalization, robustness, and failure modes.
Collaborate with Finetuning and Alignment Science teams to translate insights into production improvements.
Mentor researchers, publish findings, and grow institutional knowledge around reward modeling.
求人の全文を見る
業務内容、求める人物像、スキル、待遇 — 無料アカウントの作成で閲覧できます。
または
すでにアカウントをお持ちですか？
ログイン類似の求人
あなたに合いそうな他の職種。