Joaquín Guirao
Alicante, Valencian Community, Spain
Summary
🤩
Rockstar🎓
Top SchoolJoaquín Guirao is a software engineer with 8 years of experience specializing in C++ (C++11/14/17, Boost) and Embedded Linux, with a strong focus on computer vision and machine learning pipelines. He has contributed to high-impact open-source projects at NVIDIA and ONNX, optimizing GPU-accelerated image decoding and integrating data-loading enhancements for large-scale speech and multimodal AI frameworks. Comfortable across backend, kernel-level optimization, and ML engineering, he enjoys tackling hard problems and shaping software architecture with elegant solutions. Based in Alicante, Spain, he combines a telecoms and signal-processing academic background with hands-on GPU kernel and preprocessing work—bringing both theoretical rigor and production-oriented pragmatism to ML systems.
8 years of coding experience
Exchange Student, Informática, A, Exchange Student, Informática, A at KTH Royal Institute of TechnologyMaster of Science (M.Sc.), Telecommunications Engineering, 8.6 / 10.0, Master of Science (M.Sc.), Telecommunications Engineering, 8.6 / 10.0 at Universidad Politécnica de ValenciaBachelor of Science (B.Sc.), Sound and Image Telecommunication Engineering, Bachelor of Science (B.Sc.), Sound and Image Telecommunication Engineering at Universidad de Alicante
Polish, English, Spanish
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (24)
pytorch10
c-language10
gpgpu10
python10
artificial-neural-networks10
image-processing10
machine-learning10
onnx10
kernel10
jpeg10
deep-learning10
tensorflow10
gpu10
asr10
neural-network10
Programming languages (6)
Github contributions (5)
NVIDIA/DALI
Dec 2018 - Jan 2023
A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.
Role in this project:
Back-end & ML Engineer
Back-end & ML EngineerContributions:1 release, 2449 reviews, 458 commits in 4 years 1 month
Contributions summary:Joaquín was a core contributor to the NVIDIA DALI (Data Loading Library) project, focusing on performance optimization and feature implementation. They primarily worked on improving the image decoding capabilities, particularly for JPEG-related formats, and implementing new GPU kernels for image processing tasks, such as the JpegCompressionDistortion operator. Their work included extensive kernel development and refactoring, suggesting a strong emphasis on optimizing image processing pipelines for GPU execution, while working with various input tensor types.
pythontrainingpaddlegpu-accelerationbuilding-blocks
onnx/onnx
Oct 2021 - Aug 2022
Open standard for machine learning interoperability
Role in this project:ML Engineer
ML EngineerContributions:70 reviews, 9 commits, 11 PRs in 10 months
Contributions summary:Joaquín primarily contributed to the development of ONNX, an open standard for machine learning interoperability. Their work involved implementing new utilities, specifically related to merging models, managing prefixes, and expanding dimensions. They also addressed code review comments, fixed imports, and made various style fixes, demonstrating a focus on code quality and maintainability within the project. The user's contributions are centered around enhancing the ONNX framework's capabilities for model manipulation and interoperability.
pytorchmxnetdeep-learninginteroperabilitymachine-learning
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.