Alpha Leak is a production-grade, multi-phase trading intelligence system built for Pump.fun on Solana. This page covers what the system does, the scale it operates at, and how these docs are structured.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/0xW1re/solvedocs/llms.txt
Use this file to discover all available pages before exploring further.
What it does
Real-Time Ingestion
Processes every trade, token creation, and graduation event the moment it lands on-chain. No polling, no delays — the pipeline sees everything as it happens.
Wallet Intelligence
Builds a scored, feature-rich profile for every wallet it observes — graduation rate, win rate, entry timing, hold behaviour, return consistency — updated on a rolling 30-minute cycle across up to 5,000 wallets per pass.
Adversarial Detection
Detects bots, coordinated bundles, copy-traders, serial ruggers, wash traders, and exit-liquidity setups in real time. Adversarial signals feed directly into ML features and live position management.
ONNX ML Inference
Runs LightGBM models compiled to ONNX for sub-millisecond inference. Models are calibrated with Platt scaling and scored every 5 seconds against a 68-feature vector. Both standard and genesis models run concurrently.
Genesis Scoring
A dedicated subsystem observes every newly created token for 60 seconds, accumulates early buyer behaviour and price dynamics in memory, then runs a separate family of genesis models to identify tokens likely to 3x, 5x, or 10x shortly after launch.
Live Execution
Executes buys and sells on the Pump.fun bonding curve using atomic transaction bundles. Position sizing, strategy selection, take-profit, stop-loss, and forced exits on anti-signals are all handled automatically, with full trade history persisted to the database.
System scale
The numbers below reflect the operational depth of the system at steady state.| Dimension | Detail |
|---|---|
| Pipeline services | 30+ concurrent background services |
| ML feature vector | 68 features (standard), 75 features (genesis) |
| Wallet scoring cadence | Up to 5,000 wallets every 30 minutes |
| Bundle detection window | 5-second clustering, 10-minute interval |
| Anti-signal scan | Every 30 seconds, across all active tokens |
| Signal crowding check | Every 60 seconds, cached in the cache layer |
| Alpha decay profiling | 8 delay buckets per wallet, hourly |
| Market regime | 4 states, reclassified every 10 minutes |
| Copy-trade classification | 3 types (bot, alert, manual), 15-minute cycle |
| Data archive | Automatic archive after 60 days |
Mintlify folder structure
If you are deploying these docs to Mintlify, place files exactly as shown below. Mintlify resolves navigation paths directly from the filesystem, so the folder structure must match the paths defined indocs.json.
docs.json navigation config references each page by its path relative to the repo root, without the .mdx extension — for example, "pipeline/ingestion". Adding a new page is as simple as creating the file in the right folder and adding its path to the navigation array in docs.json.