Rug Pull Detector
A machine-learning system for flagging meme-coin rug pulls from raw chart data. The important choice was a fair tournament: every model trains on the same feature window, so deep-learning and boosted baselines can be compared honestly.
- Role
- ML Engineer
- Period
- 2025 to 2026
- Status
- Production
How the system fits together.
The calls that shaped it.
- 01
A fair model tournament
TSMixer (the winner), LSTM, Transformer, XGBoost, and LightGBM all train on the same prepared windows, so each is judged on an identical benchmark — stratified 5-fold CV with class weighting handles the heavy imbalance.
- 02
50+ features from raw ticks
Downsamples raw 1-second ticks to 5-second buckets, then layers price velocity, RSI, ROC, buy/sell pressure, three-level wavelet decomposition, rolling-window stats, and on-chain holder-concentration metrics.
- 03
A real training loop
GPU mixed precision, early stopping with patience, LR scheduling, checkpointing, and per-fold artefacts so any fold can be inspected after the run — with a full classification report, PR curve, and confusion matrix.
The interesting work isn't the stack. It's the boundaries.