Senior Software Engineer, CUDA C Libraries at NVIDIA
Dresden, Saxony, Germany
Join Prog.AI to see contacts
Join Prog.AI to see contacts
Summary
🤩
Rockstar
🎓
Top School
Michael Costa is a Senior Software Engineer with 11 years of experience specializing in CUDA C++ libraries and high-performance backend systems, currently contributing to NVIDIA's CUDA C++ Core Libraries from Dresden. He has deep expertise modernizing and optimizing GPU-accelerated projects—most notably performance and portability work on rapidsai/cudf and cuML—and has helped adapt core C++ standard library implementations across NVIDIA and Microsoft toolchains. His background in theoretical neuroscience (PhD) and a Diplom in physics underpins a rigorous, research-minded approach to algorithmic optimization and low-level concurrency, visible in contributions to libcudacxx and Thrust. Michael combines production-focused engineering with code modernization, replacing deprecated patterns and resolving cross-compiler issues to keep libraries performant and portable. He also contributes to scientific tooling like matlab2tikz, demonstrating a knack for both developer-facing infrastructure and domain-specific automation.
11 years of coding experience
3 years of employment as a software developer
Diplom, Physiker, 1,9, Diplom, Physiker, 1,9 at Universität Leipzig
Doktor (Ph.D.), Theoretische Neurowissenschaften, 1.0, Doktor (Ph.D.), Theoretische Neurowissenschaften, 1.0 at Universität zu Lübeck
[ARCHIVED] The C++ Standard Library for your entire system. See https://github.com/NVIDIA/cccl
Role in this project:
Back-end Developer
Contributions:206 reviews, 503 commits, 156 PRs in 4 months
Contributions summary:Michael primarily worked on the core C++ Standard Library implementation, focusing on atomic operations and memory management. Their contributions included removing defaulted copy constructors, adapting to changes in the C++20 standard, and addressing GCC compiler issues. The user also implemented support for `std::span` and reworked existing implementations for structured bindings, ensuring compatibility across diverse CUDA platforms and compilation environments. These efforts directly improved the functionality and portability of the C++ Standard Library for NVIDIA CUDA.
MSVC's implementation of the C++ Standard Library.
Role in this project:
Back-end Developer
Contributions:1264 reviews, 101 commits, 126 PRs in 2 years 11 months
Contributions summary:Michael contributed to the C++ Standard Library implementation within the `microsoft/stl` repository. Their work involved implementing and updating various standard library features, including those related to bounded and unbounded arrays, signed ssize, and istream_iterator cleanup. The user also worked on removing deprecated features, improving span usage, and incorporating improvements related to char8_t compatibility. These changes indicate a focus on maintaining and expanding the functionality of the C++ Standard Library.
cppc-plus-plusstandard-librarystlmsvc
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.
Request Free Trial
Michael Costa - Senior Software Engineer, CUDA C Libraries at NVIDIA