Currently Building #1

I’m currently building the technical foundations needed to contribute to AI safety research as a Research Engineer.¹

Building

• Learning the Inspect evaluation framework
• Reproducing existing AI safety evaluations
• Building my first model evaluation projects
• Publishing technical notes and project write-ups

Studying

• ARENA curriculum (self-study)
• 3Blue1Brown Deep Learning series

Reading

• AI 2027
• The Alignment Problem
• Papers on AI safety, evaluations, and interpretability

Next milestones

• Publish my first Inspect evaluation
• Publish my first paper reproduction
• Apply to the next ARENA cohort