Recent / selected writing:

• Poker is a bad game for teaching epistemics. Figgie is a better one. (blog) (twitter)
• Proper scoring rules are isomorphic to order books (blog)
• Testing 30 cows per herd for H5N1 is good enough (twitter)
• (econ theory paper) Large random matching markets with localized preference structures can exhibit large cores (journal) (preprint)
• (blog)
• (twitter)

Works in progress -- These aren't published yet, but I'm looking to have conversations about them at LessOnline!

Biotech and biosecurity:

• Personal theory of change in biosecurity
• Drug development costs can range over two orders of magnitude
• The wrong lessons from thalidomide

Incentive design:

• Impact certificates under cause area uncertainty

EA inside baseball:

• “Giving 10%” and investment gains: A proposed framework
• Charitable donations for founders, startups, and the rest of us