Responsibilities
- Define and establish measurable standards for quality in Ash conversations.
- Make hypothesis-driven decisions about Ash's behavior globally and in specific contexts, adapting quickly based on evidence.
- Collaborate with the Safety team to set boundaries for Ash and manage challenging edge-case conversations.
- Recruit, scale, and mentor a distributed team of human reviewers by setting quality benchmarks, training them on standards, and ensuring consistent calibration.
- Provide clear guidance to the conversation and machine learning teams during ambiguous situations.
