Full Transparency
No black boxes. Tradewink runs 75+ concurrent AI loops that monitor markets, generate strategies, execute trades, and improve themselves — all in real time. Here is exactly how it works.
System Architecture
From a Discord slash command to a filled order at your broker — here is the complete data flow through six architectural layers.
Discord
Slash commands & mentions
Web Dashboard
Settings & monitoring
REST API
Programmatic access
App
Composition root
WatchlistService
Ticker management
StrategyService
Strategy orchestration
75+ Concurrent Loops
Market-hours aware
Strategy Loops
Scan / Update / Exit
Event Bus
Priority async queue
Polygon
Primary market data
Finnhub + EDGAR
News & filings
WebSocket Streams
Real-time bars & trades
Claude Sonnet/Haiku
Analysis engine
Multi-Agent Teams
Debate & consensus
Self-Improver
ML retrainer + RL
Risk Manager
PDT, limits, circuit breaker
8 Brokers
Alpaca, IBKR, Schwab...
Per-User Keys
Fernet-encrypted
Concurrent Loops
Every loop adapts its speed to market conditions. Full speed during market hours, reduced cadence after hours, minimal overnight.
AutonomousAgent initialized — 75+ concurrent loops activeMarket Intelligence: Earnings / Insider / SEC / News / CongressReal-Time Monitoring: Momentum / Macro / Options / Dark PoolStrategy Engine: Scan / Update / Exit / Sell ConfidenceTrade Execution: Screen / Evaluate / Size / AI Score / ExecuteSelf-Improvement: ML Retrainer / Calibration / Prompt EvolutionPortfolio Health: Risk / Trim / Anomaly / Daily PlanningOptions Loops: Flow / GEX / Dark Pool / Max Pain / WheelCrypto Loops: Momentum / BTC Dominance / Fear+Greed (24/7)+ 30 more concurrent loops running right now...All loops market-hours aware / adaptive cadence / self-improvingScans earnings calendars, insider buys, SEC filings, dark pool prints, and news events — across every ticker.
Tracks price movements, volatility regimes, options gamma exposure, and institutional dark pool activity in real time.
Generates buy/sell strategies via AI, monitors active positions, manages trailing stops, and rotates between factor exposures.
The execution pipeline: screens candidates, evaluates with AI, sizes positions via Kelly Criterion, and executes through your broker.
Reviews every trade outcome, retrains ML models, adjusts confidence scores, and A/B tests prompt variations to get smarter over time.
Monitors portfolio balance, flags overweight positions, detects statistical anomalies, and generates daily trading plans.
Execution Pipeline
The autonomous day trading system runs this 6-step pipeline every scan cycle. Each step has safety gates, AI scoring, and risk checks.
Pipeline runs in both global broker mode (single scan for all) and per-user broker mode (individual scan/evaluate/execute per user).
Data Flow
Six institutional data feeds flow into the AI engine, get cross-validated, and execute inside your brokerage account — automatically.
Data Sources
Polygon
Real-time price & options
Finnhub
News & earnings
Unusual Whales
Dark pool & congress
SEC EDGAR
Filings & form 4s
ORATS
IV rank & vol data
Alpaca
Broker execution
AI Engine
Claude + Multi-model
Delivery
Discord DM
Private alerts
Webhook
Your own systems
Digest & alerts
REST API
Elite plan access
Tech Stack
Python 3.12+
Async-first with asyncio
Claude AI
Sonnet & Haiku multi-model
PostgreSQL
Production DB via asyncpg
Discord.py
Slash commands + Cogs
TA-Lib
Technical indicators (tiered fallback)
Prometheus
Metrics + health checks
Redis
Caching (in-memory fallback)
Sentry
Error tracking + tracing