Synapse-Quant
8マイクロサービス、リアルタイムGraphQL、TUIコックピット——生データからトレード仮説まで
17
Containers
41
Tech Indicators
TUI
Terminal Cockpit
CI/CD
GitHub Actions
Most crypto tools offer either pretty charts with weak data lineage or rich data pipes with poor trader UX. Synapse-Quant bridges both: 8 microservices handle real-time market ingestion, feature engineering, ML scoring, and RAG-powered context assembly—all accessible through a unified TUI cockpit where you can monitor markets and query the AI copilot without leaving the terminal.
The architecture is production-grade: GraphQL gateway with real-time subscriptions, dedicated ingestion workers for continuous data streaming, and a feature engine that computes signals and factors on live data. TypeScript 5.7 for the platform layer, Python 3.11+ for ML models, CI/CD via GitHub Actions. Open-source and MIT-licensed.