HiveQ
High-performance trading analytics: 20 years of OHLC data, backtesting engine
What needed solving
A quantitative trading client needed sub-second analytics over two decades of market data plus a real backtesting engine, infrastructure most stacks buckle under.
How I built it
Built a ClickHouse-powered time-series service over ~20 years of Databento OHLC data, KDB integration for quant data, and a Nautilus-Trader backtesting layer surfaced through a Next.js dashboard.
How the system fits together
Databento supplies OHLC market data at 1-second, 1-minute and 1-day resolution, with about 20 years of history held in ClickHouse for high-frequency time-series queries, fed by a Python data pipeline. KDB holds quantitative trading data, and a backtesting layer on Nautilus Trader simulates strategies against the historical bars and surfaces the results in the dashboard. Node.js APIs serve the HiveQ UI, a Next.js dashboard with better-auth sign-in, alongside HiveQ Billing for subscriptions and payments, the Jarvis CRM, and mock API services for development and testing.
Results
- About 20 years of historical OHLC market data available at 1-second, 1-minute and 1-day resolution
The details
Financial-data and trading-analytics platform: HiveQ Billing for subscriptions and payments, the HiveQ UI dashboard on Next.js with better-auth, the Jarvis CRM, a ClickHouse-powered service for high-frequency time-series analytics, KDB integration, and a Nautilus Trader backtesting engine running strategies against historical OHLC data.
- Databento OHLC market data (1s / 1m / 1d), ~20 years of history
- ClickHouse time-series analytics for high-frequency queries
- KDB integration for quantitative trading data
- Backtesting engine on Nautilus Trader
- HiveQ Billing, the Jarvis CRM and the HiveQ UI dashboard
We can build something that lasts.
Tell me what you are building. I will come back with an honest take on approach, scope and timeline, usually within 24 hours.