really good breakdown of building and scaling a prediction market market maker from scratch.
most threads talk about the strategy. this one talks about everything that breaks during testing: stale quotes, websocket desyncs, partial fills, reconciliation, adverse selection, ...
also a nice reminder that liquidity compounds. the whole strategy depended on being able to hedge against a deeper, more liquid market
nowadays anyone can spin up a bot from their laptop and learn the same market structure lessons that used to require a seat at a prop shop
The reporting on OpenAI and Sam Altman that I've been working on for the past year and a half, for @NewYorker, with @andrewmarantz: https://t.co/HEPHN4E54P