Researcher, Conceptual Reasoning Capabilities
Redwood ResearchBerkeley, CA
AI safety & policy
Full-time
AI safety & policy
Full-time
Salary
$350K – $850K
Routes to impact
Direct high impact on an important cause
Description
Research conceptual reasoning in AI by creating datasets and improving how models handle hard-to-verify, argument-based questions.
- Create datasets for philosophy, AI strategy, and AI safety reasoning tasks
- Evaluate LLM-written arguments and write detailed rubrics for assessments
- Build model scaffolds and run training experiments, depending on experience
- Develop new research ideas and communicate findings and theory of impact
- May manage researchers and relationships with AI labs and other stakeholders
This text was generated by AI. If you notice any inconsistencies, please let us know using this form.
