Order Book Engine (C++ core + Python)

This is a small personal project where I’m building a simple limit order book (LOB) matching engine in C++ and exposing it to Python via pybind11.

Why I’m doing this

• I want to learn more about quant trading / market microstructure by implementing the core mechanics myself.
• I want to work as a quant, and this is a hands-on way to practice building realistic building blocks.
• I also wanted to explore how to combine C++ (performance/engine) with Python (research/simulation/agents).

What’s inside (high level)

• C++ order book implementing price-time priority matching.
• C++ simulator wrapper that advances time and exposes market data.
• Python extension module named market_simulator (built from the C++ code).
• A small Python-side scaffold to run a simulation loop and plug in “agents”.

Project tree

order_book_engine/
├─ docs/
│  ├─ engine.md                     # Engine internals documentation
│  └─ setup.md                      # Setup and build instructions
├─ src/
│  ├─ book_implementation/
│  │  ├─ order_book/
│  │  │  ├─ order_book.cpp          # Core order book matching engine
│  │  │  ├─ order_book.hpp          # Order book interface
│  │  │  └─ types.hpp               # Order and trade type definitions
│  │  └─ simulation/
│  │     ├─ python_bindings.cpp     # Pybind11 bindings for Python
│  │     ├─ simulator.cpp           # Market simulation logic
│  │     └─ simulator.hpp           # Simulator interface
│  └─ py/
│     ├─ setup.py                   # Build configuration for C++ extension
│     ├─ simulator.py               # Main simulation runner
│     ├─ market_simulator.pyi       # Type stubs for C++ extension
│     ├─ helper/
│     │  ├─ data_types.py           # Pydantic models for orders and market data
│     │  └─ place_orders.py         # Helper to submit orders to simulator
│     └─ agents/
│        ├─ agent.py                # Base agent class for trading strategies
│        └─ random_agent.py         # Example agent with random trading behavior
├─ pyproject.toml                   # Python project metadata
└─ README.md

Quickstart (Windows)

1) Install Python deps

From the repo root:

# Create venv
python -m venv .venv

# activate
./.venv/Scripts/activate

# install dependencies
pip install .

or use uv:

uv sync

2) Build the Python extension (pybind11)

This project builds a native module called market_simulator from C++ sources.

cd src\py
python setup.py build_ext --compiler=mingw32 --inplace

After this, you should be able to import the module from Python.

3) Run the Python simulation scaffold

python simulator.py

Notes

• Setup details and prerequisites (MSYS2/MinGW): see docs/setup.md.
• More engine internals (data structures, matching logic): see docs/engine.md.