Ray Zhang
Algorithm Developer at Hudson River Trading
San Francisco, California, United States
Summary
👤
SeniorRay Zhang is an algorithm developer with 11 years of experience, currently building low-latency trading systems at Hudson River Trading after roles at Airbnb, Citadel, and Bloomberg. A UCLA computer science graduate with highest distinction (3.94 GPA), he blends rigorous math and probability insights with pragmatic C/C++ and systems engineering. Ray is an active open-source contributor—fixing JSON5 handling in JupyterLab, enhancing C++ tensor ops in xtensor, and improving Rust-based Polars serialization—reflecting a cross-language fluency from Python to Rust and C++. He focuses on algorithmic robustness, performance, and reproducible tooling, and maintains a technical blog where he writes about math and CS. Colleagues rely on him for tricky numerical and serialization edge cases that often surface in production data pipelines.
11 years of coding experience
Stackoverflow
3,778reputation
256kreached
60answers
110questions
python
top-5%
Github Skills (36)
json10
javascript10
c-language10
mstest10
datatypes10
python10
multidimensional-arrays10
dataframes10
data-serialization10
polar10
gt10
jupyterlab10
numpy10
dataframe10
serialization10
Programming languages (8)
Github contributions (5)
xtensor-stack/xtensor
Jul 2018 - Aug 2018
C++ tensors with broadcasting and lazy computing
Role in this project:
Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:15 commits, 11 PRs, 11 comments in 1 month
Contributions summary:Ray primarily contributed to improving the xtensor library by adding and refining features related to random number generation, shared expression evaluation, and data serialization. They implemented reservoir sampling and sampling with replacement functionalities within the `xrandom` module, alongside associated unit tests. Furthermore, they added support for int8 data types in the `xnpy` module, and fixed issues related to compilation and mathematical operations.
cppmpinumpypython-bindingsc-plus-plus
pola-rs/polars
Aug 2022 - Dec 2022
Dataframes powered by a multithreaded, vectorized query engine, written in Rust
Role in this project:Back-end Developer
Back-end DeveloperContributions:1 review, 4 commits, 4 PRs in 3 months
Contributions summary:Ray contributed to bug fixes and feature implementations related to Polars' data type handling and serialization capabilities. They resolved issues with datatype pickling in the Python interface, specifically focusing on the `datatypes.py` and `tests/test_datatypes.py` files. The user also implemented Serde support for the `Duration` type and added cloudpickle for serializing Python UDFs, enhancing the library's usability and data handling functionality. Further contributions include removing a debugging statement and improving the output types of cumsum and cumprod.
polarsdataframespythondataframerust
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.