How I Put a Rocket Engine into Python: Porting ClickHouse for High-Performance App PythonAsia 2026

Practical techniques for embedding C++ libraries into Python using tools like PyBind11
How to handle performance, threading, and memory management in hybrid Python–C++ systems
How to profile and optimize data-intensive Python modules that rely on native code

Developers familiar with C++-based Python libraries (NumPy, Pandas, PyTorch)
Engineers curious about pushing Python’s performance limits
Anyone who’s ever wondered what happens when you try to fit a rocket engine inside a Python module 🚀

TLDR: What if you could bring the blazing speed of a C++ analytics engine directly into Python — without giving up the simplicity that makes Python so great?
In this talk, I’ll share how I ported ClickHouse, a high-performance C++ OLAP engine, into a native Python module. You’ll see the real-world challenges behind bridging these two worlds — from managing memory across language boundaries to overcoming the challenges of integrating Jemalloc in shared libraries and enabling zero-copy reads/writes Pandas DataFrames — and how I turned a massive C++ system into something that could be imported with a simple import statement.

Category: Data Science/Analytics/Engineering Audience Level: Advanced

Auxten
- 👨🏻‍💻 Experience in RecSys, Database
- Technical Director of ClickHouse core team
- Principal Engineer in Shopee (ML Platform)