Introduction

​

What is asset-pricing?

An end-to-end market intelligence platform for Indian equities. Kite Connect crawlers pull every NSE instrument and 25 years of daily candles into Postgres; financial statements become point-in-time metrics; predictor panels feed an ML pipeline that ranks stocks monthly; QuantConnect Lean turns those ranks into auditable backtests measured against the Fama-French six-factor benchmark.

​

What makes it careful

• Point-in-time correctness — a metric value for any month uses only data knowable then. Statement-backed metrics respect a configurable reporting lag (asof_lag), so no lookahead bias leaks into features or backtests.
• One metric registry — every predictor is a registered function with a standard financial category (for humans) and an ML feature-group alias (for models). One decorator wires it into panels, API, and docs.
• One API, two paths — a Python library holds all logic; the REST layer is a thin 1:1 wrapper with zero logic of its own.

​

Where to start

• Quickstart — set up, crawl, build panels, run the API
• Architecture — the four layers
• Metric registry — the taxonomy and how to add a predictor
• API reference — library and REST, side by side