about 2 hours ago
Base Salary
$295k - $445k/yr
Responsibilities
- Develop a deep understanding of effective collaboration traits in agents.
- Translate qualitative judgments into hypotheses, evaluations, and training interventions.
- Analyze user signals to identify behaviors that foster trust and satisfaction.
- Collaborate with experts to produce high-quality training data.
- Enhance reward models and reinforcement learning objectives.
- Work with pretraining teams on data and objectives that influence personality.
- Establish sustainable pipelines for updating training data.
- Partner with product teams to translate consumer insights into model improvements.
Requirements
- Strong user-centric perspective on model interactions.
- Ability to convert subjective product questions into testable hypotheses.
- Commitment to preserving individuality and behavioral diversity in models.
- Technical foundation in machine learning, software engineering, or related fields.
- Strong taste for model behavior and user feedback interpretation.
- Experience with LLMs, post-training, and reinforcement learning.
- Ability to navigate ambiguous problems with noisy signals.
- Effective communication skills across diverse teams.
Categories
AI & MLData Science