Turn any prediction into ranked Kalshi/Polymarket trades [video]
Author calls it 'super shitty V0 MVP'—embryonic but the prediction-to-trade angle is real.

Shows more than raw P&L: six metrics, trader archetypes (DIRECTIONAL, ACCUMULATOR, ALGO_TRADER), persistent grades and filters (active/inactive, search by wallet) make it easy to scan for repeat winners. The UX puts useful columns up front — volume, predictions, win rate and a computed grade — but I want to see the methodology and survivorship controls up front; leaderboard data without clear adjustments can mislead casual users.
Polymarket users, prediction‑market traders, crypto traders, data‑driven investors and researchers
Author calls it 'super shitty V0 MVP'—embryonic but the prediction-to-trade angle is real.
Single-screen dashboard pulls a Polymarket wallet by address and surfaces positions, PnL, allocations and trade history — exactly what traders need when juggling bets across markets. The UI is deliberately minimal and readable, focusing on quick wallet lookups rather than bloated features. Useful as a lightweight explorer, though it would earn higher marks with wallet-connect, exports or deeper filtering.
Gamifies commit counts, but GitHub's own contribution graph already solves this.
Uses KL divergence to give asymmetric moves the weight they deserve (a 5%→10% jump reads way louder than 50%→55%), then multiplies that by log-volume, historical SNR and trajectory consistency to surface genuinely interesting shifts. Practical engineering choices — persisted snapshots, rolling detection windows, cooldown dedupe and Telegram delivery — make this something you can drop into a trading workflow and actually rely on.
Elo leaderboard for game nights, but dozens of free Discord bots do exactly this.
GitHub commit leaderboard; removes novelty once the initial curiosity wears off.