Adam Doerr

The @Firelightfi dream isn’t just DeFi insurance. The dream is programmable insurance. With @Firelightfi, we obviously have plenty of very practical, near-term goals: making DeFi cover actually work at scale, giving LPs a sane risk/return profile, and protecting blue-chip protocols in a credible way. That alone will keep us busy for a while. But the long-term ambition is bigger: use web3’s full expressive power to build a new kind of insurance primitive. Not just “on-chain paperwork,” but insurance that is composable, granular, and programmatically triggered. Most insurance workflows today are completely static. You negotiate a bespoke policy, lock it in, and if there’s a claim it goes through a months-long manual review. Now imagine those workflows on-chain. @Firelightfi starts by covering four core classes of DeFi events: Technical exploits Economic exploits Oracle manipulations Bridge events Then comes scope. From a DeFi perspective, Firelight thinks in layers: Protocol → Assets → Markets → Transactions The deeper you go, the more interesting programmability becomes. At protocol level, it’s basically a big umbrella. At market/pool level, you can start to tailor behavior. At transaction level, things get wild. Picture a user bridging assets, swapping on the destination chain, and bridging back. The source dApp could talk to Firelight, insure all three transactions before execution, and settle the premium after completion — flash-loan style flows, but for risk. We’re not there yet with @Firelightfi . But that’s the direction of travel: From “DeFi cover” → programmable insurance as a first-class web3 primitive.