Essays · Operations Intelligence, AI and Quant
Hossein Narimani — Writing
In-depth writing on quant system design, operational AI, SaaS architecture, custom AI agents and founder execution systems.
OHLCV Feature Pipeline Architecture: Designing Feature Registries and Caches for Scalable Trading Systems
Most trading systems do not fail because of the model. They fail because every component interprets market data differently.When a feature has one value in a backtest and another in live trading, the problem is not intelligence. It is architecture.The Real...
Read more →
OHLCV Pump Pattern Analysis: Detection, Limits, and Alert System Design
OHLCV Pump Pattern Analysis: Detection, Limits, and Alert System DesignThe hard problem is not finding a green candle.The hard problem is...
Read more →
OHLCV Data Quality Checks: What to Validate Before Backtesting or Trading
Most backtests do not fail because of the model. They fail because of the data.If your OHLCV feed is incomplete, inconsistent, or retrospectively...
Read more →
Deep Learning Prediction Models for Stock OHLCV Data: A System Design Framework
The Real Problem: Why Deep Learning on OHLCV Data Usually DisappointsMost teams building deep learning models for stock prediction start with the...
Read more →
Quant Research vs Quant Trading: Signal Discovery vs Capital Execution in Quantitative Systems
Most people treat Quant Research and Quant Trading as different labels for the same profession. Operationally, they are two separate layers of the...
Read more →