alphabench.in
AI agents for quantitative research on Indian equity markets. Describe a trading strategy in plain English and get a full institutional-grade backtest in seconds.
Live Platform
What It Is
alphabench is the first multi-agent quantitative research platform purpose-built for Indian equity markets. Retail traders and institutional quants face the same bottleneck: turning a raw strategy idea into a backtested, production-ready system requires weeks of data wrangling, Python scripting, and manual analysis.
alphabench collapses that workflow into a conversation. Users describe their strategy in plain English. The platform's AI agent system handles intent parsing, data retrieval, backtest execution, and performance diagnostics, returning results in under a minute. Live trading connects directly via Zerodha Kite, and paper trading simulation lets strategies run in real time before any capital is committed.
The Agent System
Planner
Parses plain-English strategy descriptions and orchestrates the research pipeline end-to-end.
Quant
Compiles and executes backtests on RaptorBT, a Rust-based engine for deterministic, reproducible results.
Researcher
Discovers stocks via fundamental screening across 1,800+ NSE & BSE equities with 10+ years of historical data.
Diagnostician
Analyses trade-level performance, surfaces drawdown drivers, and suggests iterative refinements.
Data & Infrastructure
1,800+ NSE & BSE stocks with 10+ years of historical daily OHLCV data and 1-minute intraday candles
30+ fundamental fields per equity for screening: P/E, ROCE, debt ratios, promoter holding, and more
Options chain data enabling multi-leg derivatives strategy testing across NSE F&O
RaptorBT: a Rust-based open-source backtesting engine built for deterministic, reproducible results at speed
WebSocket-driven real-time P&L monitoring once strategies go live via Zerodha Kite integration
Platform Scale
Strategies Backtested
Data Points Analysed
NSE & BSE Stocks Covered
Want us to build yours?
alphabench.in is what we build when we have the full stack to ourselves. If you need a data-intensive platform, AI agent system, or trading infrastructure, let's talk.