Polysemantic human

I build AI systems for a living and for fun. I'm particuraly obsessed with discovering ways these systems fail.

Over the years, I've worked as a practitioner, a researcher, and a people leader. In whatever capacity, I have enjoy sinking into hard and curious problems. Lately I'm been fascinated with the intersection of LLMs and psychology both in explaining the strange ways LLMs behave from nuerobiology perspectives, and better understanding how LLMs change the way people behave too.

Writing is nature's way of telling us how lousy our thinking is.

— Leslie Lamport

I like to write about whatever I find interesting. This is where I write to fomulate my own ideas on a topic. By writing, my primary goal is to expose lousy thinking from self-reflection and reckon with alternative viewpoints (hopefully from well-informed internet strangers!... See xkcd comic on posting to the internet)

If any of this is interesting to you, I do hope you reach out to say hello :)

Experience

• Vanguard
  • Head of AI Safety & Evaluation
  • Principal Data Scientist
  • Senior Data Scientist
• Novodynamics
  • ML Scientist
• Bank of America
  • Data Science Consultant
• American College of Radiology
  • Data Scientist
• New Haven Public Schools
  • Physics Teacher

Interests

• AI Evaluation
• Adversarial ML methods
• AI red teaming
• Software engineering leadership
• Health & Fitness
• Behavioral Economics
• Board games
• Vegan Cooking
• Data journalism & information design
• Tiny projects
• Technical communication
• Education

Disclaimers

The views expressed here are my own and do not reflect any employer or client.