Ginés Martínez
Staff Research Engineer At Reality Labs Research at Meta
Pittsburgh, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top School
Top expert inDeep Learning and Computer Vision Technologies
Ginés Martínez is a Staff Research Engineer at Reality Labs Research (Meta) with nine years of experience turning cutting-edge perception research into production-ready systems. He is the lead author of OpenPose, the widely-cited, 25k+ star real-time whole-body keypoint detection library, and has driven production ML inference work such as UE5’s Neural Network Inference at Epic and integrations with ONNX Runtime. Ginés combines deep C++ and systems expertise—optimizing CPU rendering, multi-threading, and cross-platform compiler compatibility—with hands-on ML training and dataset engineering for pose estimation. He mentors engineers, designs APIs, and bridges collaboration between research teams and industry partners like Microsoft. Based in Pittsburgh, he brings a rare mix of research rigor and pragmatic engineering that reliably scales prototypes into deployed tooling.
9 years of coding experience
7 years of employment as a software developer
Engineer’s Degree Telecommunications Engineering, Engineer’s Degree Telecommunications Engineering at Universidad Politécnica de CartagenaEngineer’s Degree Telecommunications Engineering (Erasmus fellowship), Engineer’s Degree Telecommunications Engineering (Erasmus fellowship) at FH St. Pölten – University of Applied SciencesMaster’s Degree Master of Science Program Robotics Institute, Master’s Degree Master of Science Program Robotics Institute at Carnegie Mellon University
Spanish, English
Stackoverflow
777reputation
201kreached
26answers
2questions
Github Skills (34)
c-language10
operation10
scripting10
directml10
tensorrt10
machine-learning10
onnx10
data-management10
script10
deep-learning10
trainings10
tensorflow10
sh10
computer-vision10
tensor10
Programming languages (4)
Github contributions (5)
CMU-Perceptual-Computing-Lab/openpose
Apr 2017 - Oct 2022
OpenPose: Real-time multi-person keypoint detection library for body, face, hands, and foot estimation
Role in this project:
Back-end Developer
Back-end DeveloperContributions:19 releases, 521 commits, 231 PRs in 5 years 6 months
Contributions summary:Ginés's contributions primarily focused on modifying and extending the C++ OpenPose library. They added Doxygen documentation for the wrapper class and updated functions related to face and hand rendering. They also worked on improving the CPU rendering speed and fixed bugs within the core modules. The commits demonstrate work with core OpenPose functionality and contributions that likely enhanced its usability and performance.
multi-personhuman-behavior-understandingreal-timedetection-libraryhuman-pose
CMU-Perceptual-Computing-Lab/openpose_train
Aug 2019 - Nov 2021
Training repository for OpenPose
Role in this project:ML Engineer
ML EngineerContributions:22 commits, 3 PRs, 15 pushes in 2 years 3 months
Contributions summary:Ginés contributed to the training process for the OpenPose project. Their work involved modifying training scripts, including downloading and integrating datasets for body, foot, face, hand, and dome pose estimation. They added experimental models and updated the training pipeline, including solver configurations and bash scripts. These changes indicate a focus on model training and dataset management, core components of the OpenPose project.
pytorchiccv-2019deep-learningcomputer-visionmachine-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.