stereo
Machine Learning Engineer
Summary
🤩
Rockstarstereo is a Machine Learning Engineer with 12 years of hands-on experience building high-performance C++ systems for computer vision and ML. They contribute to prominent open-source projects like mlpack and tiny-dnn, optimizing core algorithms (notably accelerating SparseAutoencoder) and improving code quality across header-only C++ libraries. Comfortable across model development, serialization, testing, and back-end optimizations, stereo pairs deep algorithmic understanding with pragmatic engineering to ship reliable, efficient implementations. Their work shows an eye for maintainability—suppressing type-cast warnings and tightening examples—helping projects stay production-ready. Equally at home with Qt5 app development, they bridge research-grade models and deployable software with a systems-oriented mindset.
12 years of coding experience
Stackoverflow
4,991reputation
725kreached
50answers
116questions
image-processing
top-5%
opencv
top-1%
computer-vision
top-5%
Github Skills (28)
algorithm10
code-optimization10
opencv10
c-language10
optimizers10
machine-learning10
deep-learning10
optimisation10
neural-network10
cprogramming-language10
optimization10
image-processing9
rms9
computer-engineering9
computer-vision9
Programming languages (9)
Github contributions (5)
mlpack/mlpack
Sep 2015 - Mar 2017
mlpack: a fast, header-only C++ machine learning library
Role in this project:
ML Engineer
ML EngineerContributions:253 commits, 37 PRs, 5 pushes in 1 year 6 months
Contributions summary:Stereo's contributions primarily focused on improving the performance and flexibility of the `SparseAutoencoder` class within the `mlpack` library. They enhanced the speed of the `SparseAutoencoder` through code optimizations. The user made multiple refinements to the example code, and ensured the examples' basis size was appropriate. Further contributions included adding tests for the modified `SparseAutoencoder` implementation, and adding serialization/deserialization capabilities to persist the trained model.
regressionheaderdeep-learningscientific-computingc-plus-plus
tiny-dnn/tiny-dnn
Feb 2016 - Feb 2016
header only, dependency-free deep learning framework in C++14
Role in this project:Back-end Developer
Back-end DeveloperContributions:10 commits, 4 PRs, 14 comments in 20 days
Contributions summary:Stereo primarily focused on suppressing type cast warnings within the C++ code, suggesting a focus on code quality and adherence to coding standards. Their contributions included modifications across multiple files, particularly related to the `tiny_cnn` framework, which involved updating optimizers, layers, and network functionalities. The user also worked on integrating code changes, fixing type cast warnings, and merging updates to improve overall code stability.
cppheaderdeep-learningc-plus-plusmachine-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.