Alexander Ocsa
Senior Solutions Architect, Data Processing
São Paulo, Brazil
Summary
🤩
Rockstar🎓
Top SchoolAlexander Ocsa is a Staff Software Engineer and PhD-trained systems specialist with a decade of experience building GPU-accelerated query engines, high-performance numerical kernels, and distributed data-processing libraries. He architects parallel software that squeezes performance out of modern accelerators and memory hierarchies, including spill-aware execution nodes spanning GPU HBM, RAM, and disk. His contributions to flagship open-source projects like Apache Arrow and NVIDIA RAPIDS (cuDF) show a focus on core compute kernels and query-engine primitives—work that improves in-memory analytics and GPU DataFrame reliability. At Voltron Data and BlazingDB he designed novel distributed algorithms and task-splitting strategies that materially boosted scalability and robustness for join and count-distinct workloads. He combines academic rigor from a PhD with hands-on systems engineering across CUDA, C++, and large-scale distributed I/O, and has applied GPU HPC techniques to domains as varied as climate research and deep learning.
10 years of coding experience
12 years of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at USP - Universidade de São PauloDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Universidad Nacional de San Agustin de Arequipa
English, Portuguese
Github Skills (16)
parquet10
apache-arrow10
data-structures10
cuda10
unit-testing10
dataframes10
gpu-programming10
c-language10
cprogramming-language10
data-structure10
data-processing10
dataframe10
testing10
algorithms9
data-analysis9
Programming languages (8)
Github contributions (5)
NVIDIA/cudf
Jun 2018 - Oct 2019
cuDF - GPU DataFrame Library
Role in this project:
Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:5 reviews, 90 commits, 15 PRs in 1 year 3 months
Contributions summary:Alexander implemented unit tests to validate the `filterops` functionality within the `cudf` library. They developed new test cases, specifically targeting augmented filter operations with different data types. These tests likely contribute to ensuring the correctness, reliability and stability of the GPU DataFrame library, verifying core comparison operations. The user also incorporated test cases for the `gpu_concat` function.
cudfdataframegpurapidsarrow
apache/arrow
Aug 2021 - Oct 2021
Apache Arrow is the universal columnar format and multi-language toolbox for fast data interchange and in-memory analytics
Role in this project:Back-end Developer
Back-end DeveloperContributions:91 reviews, 7 commits, 8 PRs in 1 month
Contributions summary:Alexander primarily worked on implementing and testing features within the Apache Arrow codebase, focusing on the compute module. Their contributions involved implementing a Union ExecNode and Drop Null kernels, demonstrating a focus on data processing and transformation capabilities. The user also contributed to adding a SelectKSinkNode, integrating the SelectK kernel into the query engine for optimized data retrieval. These efforts suggest a focus on enhancing the core functionality and performance of the data processing framework.
apache-arrowarrowparquet
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.