Christian Convey
South Kingstown, Rhode Island, United States
Summary
🤩
Rockstar🎓
Top SchoolChristian Convey is a seasoned software engineer with 11 years of experience specializing in deep learning compilers, performance optimization, and hardware enablement across companies like OctoML and Intel. He excels at tackling open-ended optimization problems and has a strong track record debugging and tuning full-stack AI pipelines from frameworks down to hardware interfaces. His open-source contributions include substantive work on the Apache TVM ONNX importer and performance-focused benchmarks, and build/tooling improvements for Intel's nGraph project. Comfortable in C/C++, Python, and build systems, he combines low-level systems expertise with pragmatic DevOps and automation skills. Based in Rhode Island, he also brings program leadership experience from defense and academia, having led cross-platform scientific and simulation software projects. Colleagues rely on him to push both code performance and team effectiveness in challenging, performance-critical environments.
11 years of coding experience
27 years of employment as a software developer
Master of Science (ScM), Computer Science, Master of Science (ScM), Computer Science at Brown UniversityBS, Dual major: Computer Science, Mathematics, BS, Dual major: Computer Science, Mathematics at Gordon CollegePhD, Computer Science, PhD, Computer Science at University of Rhode Island
English, German
Stackoverflow
1,242reputation
56kreached
5answers
11questions
Github Skills (21)
c-language10
python10
machine-learning10
bash10
cmake10
hexagons10
deeplearning-ai10
compiler-compiler10
deep-learning10
build-automation10
compiler10
cprogramming-language10
code-formatting10
cicd9
onnx9
Programming languages (4)
Github contributions (5)
apache/tvm
Oct 2021 - Nov 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
ML Engineer
ML EngineerContributions:155 reviews, 35 commits, 36 PRs in 1 year
Contributions summary:Christian contributed significantly to the ONNX importer within the TVM deep learning compiler stack. Their work included adding importer code for the QLinearMatMul operator, enabling unit tests and addressing limitations. Furthermore, the user added a unit test exploring Hexagon HVX usage, suggesting involvement in performance optimization. The user also refactored benchmark-table code and added a max_pool2d benchmark.
metalvulkancompilertensoropencl
NervanaSystems/ngraph
Sep 2017 - Oct 2019
nGraph - open source C++ library, compiler and runtime for Deep Learning
Role in this project:DevOps Engineer & Build Engineer
DevOps Engineer & Build EngineerContributions:15 commits, 5 PRs, 14 pushes in 2 years
Contributions summary:Christian primarily focused on improving the build and formatting processes within the nGraph repository. They added version checking for the clang-format tool and created bash scripts to automate code formatting. Additionally, they updated the CMake configuration to apply the NGRAPH_TARGET_ARCH setting to the MKLDNN build, and addressed compile errors. The user also reverted a change related to building MKLDNN due to issues with parallel builds and make jobserver warnings.
inference-enginecppc-librarydeep-learningtvm
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.