What it is
AlphaCloud is a fully automated trading bot for Kalshi's temperature markets. It trades contracts like "Will the high in NYC tomorrow be between 45 and 46°F?" — binary bets priced from 1¢ to 99¢ that pay out $1.00 if they're right.
20 cities. 27 market series. 300+ markets per scan. 12 scans per day.
The live dashboard shows real-time paper trading performance across 13 parameter configurations.
Why this exists
This is the data scientist side of me wanting a playground. Weather prediction markets are a fascinating niche — the underlying data is completely public (weather forecasts), the markets are liquid enough to trade, and there's real edge available if you can blend models better than the average participant.
The forecast pipeline
Every 2 hours, the bot pulls ensemble forecasts from four weather model sources — 123 members total:
- GEFS — 31 ensemble members from NOAA (US)
- ECMWF — 51 ensemble members from the European Centre (generally considered the best in the world)
- ICON — 40 ensemble members from Germany's DWD
- NBM — 1 deterministic member from the NWS National Blend of Models
All 123 members are pooled into a Gaussian distribution per city per day, then calibrated with EMOS — per-city learned coefficients that correct for systematic bias. The calibration delivers ~49% improvement in forecast accuracy across 15 cities.
Edge detection and execution
The edge detector converts ensemble probabilities into bracket prices using Gaussian CDF with half-integer settlement boundaries, then compares them against what Kalshi's market is actually pricing.
If my models say 40% and the market says 25¢ — that's a potential edge.
Position sizing uses 3-level adaptive Kelly criterion: per-bet uncertainty adjustment, within-event multi-outcome optimization, and cross-event portfolio caps. Execution runs an IOC/GTC hybrid — high-confidence trades execute immediately, lower-confidence trades sit as limit orders.
Self-improving ML
This is the part I'm most excited about. The system doesn't just trade — it learns from every trade and automatically gets better over time.
- Data collection — every scan stores raw ensemble members, orderbook snapshots, edge evaluations, and forecasts (~1MB/day)
- Drift detection — rolling Brier scores with CUSUM test detect when model calibration degrades
- Auto-retrain — when drift is detected, EMOS calibration, edge classifier, and probability calibrator retrain automatically from accumulated data
No manual intervention needed. The more it trades, the smarter it gets.
Where it's at
Paper trading on Railway — running the full pipeline against real Kalshi markets with simulated money. 13 parameter configurations test different edge thresholds and sizing aggressiveness simultaneously. First settled P&L came in on Day 1 — the best config (10% edge, 25% Kelly cap) is profitable.
Go-live target is April 1, 2026 with $250 real capital — if paper results stay within 30% of backtest projections through March 31.