Grant Jensen
Software Engineer at Google
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolGrant Jensen is a software engineer in San Francisco with six years of experience focused on high-performance on-device machine learning and GPU inference. At Google he has driven TensorFlow Lite CPU and GPU performance improvements, contributing optimized kernels and XNNPACK integrations to the flagship tensorflow/tensorflow and google/XNNPACK projects. His background spans full ML stacks—from building GPU-accelerated portfolio and NLP models at NVIDIA to end-to-end ML systems and cloud integrations at Red Hat—rooted in a BS in Mathematics and Computer Science from UW–Madison. Known for obsessing over fast code and subtle performance wins, he pairs rigorous test automation with kernel-level optimization to make ML run faster on constrained devices.
6 years of coding experience
3 years of employment as a software developer
Bachelor of Science - BS, Mathematics and Computer Science, Bachelor of Science - BS, Mathematics and Computer Science at University of Wisconsin-Madison
Github Skills (18)
c-language10
float3210
testing10
xpack10
tensorflow10
performance-optimization10
floating-point10
tflite10
cprogramming-language10
test-automation10
machine-learning9
inference9
deep-learning8
algorithm7
algorithms7
Programming languages (4)
Github contributions (5)
tensorflow/tensorflow
Jul 2022 - Jan 2023
An Open Source Machine Learning Framework for Everyone
Role in this project:
ML Engineer
ML EngineerContributions:25 reviews, 18 commits, 2 PRs in 6 months
Contributions summary:Grant primarily contributed to the TensorFlow Lite (TFLite) library, focusing on kernel implementations and optimizations. Their work involved integrating XNNPACK for performance improvements in various operations like transpose, multiplication, and activations. The contributions also included fixing bugs, updating build scripts, and adding support for new features like the PadV2, Select, and Gather operations, indicating a focus on enhancing TFLite's capabilities and performance.
pythondata-sciencedeep-learningmlmachine-learning
google/XNNPACK
Aug 2022 - Oct 2022
High-efficiency floating-point neural network inference operators for mobile, server, and Web
Role in this project:Test Automation Engineer
Test Automation EngineerContributions:14 commits in 1 month
Contributions summary:Grant added and modified numerous test cases within the `google/xnnpack` repository. Their work primarily involved creating and refining tests for the eager APIs of various floating-point operators, including transpose, ceiling, clamp, ELU, hardswish, leaky ReLU, negate, sigmoid, square, square root, and truncation. The changes show a strong focus on verifying the functionality and correctness of these operators across different dimensions and configurations. The user also contributed to adding and modifying header files and build files related to testing these functions.
multithreadingsimdtensorflowcpufloating
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.