Percy Aucahuasi is a Senior Software Engineer with 18 years of experience building high-performance, data-intensive systems across C++, Java, Python and GPU-accelerated stacks. He combines hands-on kernel and datetime work in prominent open-source projects like Apache Arrow and RAPIDS cuDF with product-focused engineering at companies such as Graphistry and BlazingSQL, where he integrated GPU SQL engines with cloud and distributed tools. Percy has led cross-functional teams, modernized build and CI systems, and fixed subtle race and memory issues—skills reinforced by contributing novel scalar functions and temporal kernels to Apache Arrow. His MBA complements deep technical expertise, enabling him to bridge strategy, product and engineering while championing open standards like Substrait. Based in Peru, he brings a rare mix of low-level performance tuning and pragmatic system design for analytics at scale.
18 years of coding experience
12 years of employment as a software developer
Computer Software Engineering, Diploma course, Computer Software Engineering, Diploma course at Universidad de San Martín de Porres
Master of Business Administration - MBA, Strategy, Leadership, Finance, Marketing, Operations and Social Responsibility, Master of Business Administration - MBA, Strategy, Leadership, Finance, Marketing, Operations and Social Responsibility at CENTRUM PUCP
Bachelor's degree, Computer Software Engineering, Bachelor's degree, Computer Software Engineering at Universidad Peruana de Ciencias Aplicadas
Contributions:47 commits, 4 PRs, 28 comments in 1 year 8 months
Contributions summary:Percy's commits primarily involve modifications to the `cudf` library, specifically focusing on datetime operations. They implemented and tested datetime-related functions, including extracting year, month, day, hour, minute, and second from date64 and date32 data types. The changes encompass modifications to `datetimeops-test.cu` and `datetimeops.cu` files, demonstrating a focus on extending the functionality and ensuring the correctness of date and time handling within the GPU DataFrame library. The commits include error handling and size validation for added robustness.
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
Contributions:132 reviews, 10 commits, 15 PRs in 1 year 4 months
Contributions summary:Percy primarily contributed to the Apache Arrow project by implementing and improving C++ kernels, specifically related to data processing functionalities like "Take" and temporal component extraction. They addressed bug fixes and implemented new features for existing kernels, improving their handling of edge cases and expanding their capabilities. Furthermore, the user contributed to the project's documentation, including minor fixes. They also implemented new scalar functions, such as count distinct and list element, adding to the projects function library.
memorymulti-languagetoolboxacceleratedarrow
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.