Skip to main content

NanoARB

Nanosecond-level high-frequency trading framework for CME futures markets

Build production-grade trading strategies in Rust with sub-microsecond inference latency, ML-powered signals, and realistic backtesting.

Get startedArchitecture

Sub-microsecond latency

Achieve <800ns tick-to-trade latency with production Rust codebase and zero Python at runtime.

ML-powered strategies

Mamba State Space Models 10-50x faster than Transformers with RL-based market making.

Realistic backtesting

Event-driven engine with latency simulation, queue position modeling, and adverse selection.

Get started

Follow these steps to build and run your first trading strategy with NanoARB.

1

Install Rust and dependencies

NanoARB requires Rust 1.75+ and optionally Node.js for the dashboard UI.
# Install Rust
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

# Clone the repository
git clone https://github.com/dhir1007/nanoARB.git
cd nanoARB
2

Build and run the engine

Use the provided start script to build the Rust engine and launch the dashboard.
# Build and start everything
./start.sh
This starts the trading engine at http://localhost:9090 and the dashboard at http://localhost:3000.
3

Run your first backtest

Execute a backtest with synthetic market data to validate your setup.
# Run backtest via CLI
cargo run --release --bin nanoarb -- --backtest

# Or via API
curl -X POST http://localhost:9090/api/backtest \
  -H "Content-Type: application/json" \
  -d '{"symbol":"ES","initialCapital":1000000,"useML":true}'

{
  "total_pnl": 42850.75,
  "sharpe_ratio": 4.8,
  "max_drawdown_pct": 5.2,
  "win_rate": 0.543,
  "total_trades": 48231
}

Explore the framework

Learn about NanoARB’s modular architecture and core concepts.

Architecture overview

Understand the event-driven architecture and crate structure

Order books

Learn how NanoARB maintains and updates 20-level order books

Building strategies

Implement custom trading strategies using the Strategy trait

Feature extraction

Extract market microstructure features for ML models

API reference

Comprehensive documentation of NanoARB’s public API surface.

Core types

Price, Quantity, Side, Timestamp, and OrderId types

Strategy API

Strategy trait and market maker implementations

Backtest engine

Event-driven backtesting with realistic fill simulation

Order book API

OrderBook construction and update methods

Configuration

Trading, risk, and latency configuration options

Features

Microprice, OFI, VPIN, and imbalance calculations

Ready to build?

Start developing high-frequency trading strategies with NanoARB’s production-grade framework.

Get started now