snipe.exchange / about · why we built Snipe · 2026
Two legs is one too many.
dear trader,
For years, expressing a range view on a prediction venue meant buying two opposing strikes across two markets. Two fees. Two slippages. Two positions to babysit. Capital trapped on both sides. The math worked, sometimes — the execution always cost the edge.
DeepBook Predict is the missing primitive on Sui: a shared vault that prices contracts on demand, with continuous strikes and native range markets. One contract, one fee, one position. The market structure we’d been hand-rolling — finally specified as a protocol.
We built Snipe because someone had to build the pro frontend for it.
What we believe
Risk is a first-class state. Liquidation distance, max loss, fees, slippage, break-even — every number surfaces in the trade ticket before you sign. Hidden risk is a UI failure, not a feature.
The terminal is the product. Pro traders convert on terminal feel, not on landing-page polish. Hotkeys, mono numerals, hairline density, sub-second confirmations. Marketing should look like the tool, not the tool like the marketing.
Honesty over decoration. No fabricated metrics. No “trusted by N,000+ teams.” No celebratory toasts on a loss. Every quantitative claim is real, labelled, or marked “to confirm.” Pre-mainnet, that means most of the proof bar is honest placeholders. Post-mainnet, we’ll fill them in with data, not promises.
Protocol, not middleware. Market data, positions, history, vault state — all read directly from DeepBook’s on-chain contracts and public infrastructure. Less off-chain state means less trust assumed.
What we won’t ship
40× leverage on day one
Not on a market that doesn’t yet have the depth to support it. Tiers ramp as liquidity and risk parameters validate.
“Audited” as a badge
Not until the audits land.
“Mainnet live”
Not until it is. Until then: a status page that’s accurate to the minute, and a testnet label that doesn’t hide.
If that sounds slow, it is. The alternative — overpromising, then walking it back — has cost prediction markets two cycles of trust. We’d rather move at the pace of what we can defend.
An honest disclosure
This page is not yet signed. Snipe is in testnet alpha. The team’s public bylines will appear here once the alpha cohort closes and the protocol audit completes. Until then, treat this as a letter from the team behind Snipe — anonymous on purpose, because the work matters more than the names attached to it.
We’ll update this page when names go on it. We read every alpha application.
— the team behind Snipe
snipe.exchange · sui-native · powered by deepbook · 2026