Iosif Gershteyn, Harry Crane & Philip Maymin: Djinn Subnet 103, Sports Betting Signals | Ep 96

In this episode, we are joined by Iosif Gershteyn, Harry Crane, and Philip Maymin of Djinn (Subnet 103), who detail the development of a decentralized protocol enabling sharp sports bettors to sell their signals to executing counterparties while preserving signal secrecy through multi-party computation. They discuss the sharp bettor problem of winning bettors getting limited or banned by sportsbooks, the logistical ceiling of personal betting account networks, and how Djinn decouples intelligence from execution by connecting signal generators to the 500 million dormant sports betting accounts worldwide. Iosif, Harry, and Philip also detail the architecture of Djinn's miner and validator layers, the warranty and collateral system for signal quality guarantees, the TLS Notary attestation mechanism for proving line availability, and Djinn's broader applicability to verifiable execution problems beyond sports betting. *In this episode, you'll learn* - The sharp bettor problem — how training profitable sports bettors at analytics.bet led to the discovery that winning bettors get limited or banned, motivating a protocol to separate signal generation from execution - The protocol architecture of Djinn (Subnet 103), including the roles of geniuses (signal generators), idiots (executors), miners (line attesters via TLS Notary), and validators (secret keepers via multi-party computation) - The off-chain revenue model drawing from the $1–2 trillion global sports betting market, with a 0.5% protocol fee creating an economic engine independent of Bittensor emissions - How the warranty and collateral system prevents both-sides shill attacks by requiring signal sellers to put skin in the game, and how Shamir secret sharing enables validators to grade outcomes without learning the true signal index - The 10-signal audit window mechanics, including USDC reimbursement up to the signal purchase price and Djinn credit refunds beyond, and why bet sizing should match signal notional to m

Read the article at Ventura Labs (YouTube) ↗

Taosis indexes announcements, articles, posts and releases about Bittensor subnets from their own channels and keeps a record of each. The text above is the source’s; the figures are read from the chain by Taosis.