Justin Yip
Data Janitor
Sunnyvale, California, United States
Summary
🤩
RockstarJustin Yip is a seasoned back-end engineer and "data janitor" with 12 years of experience building and hardening ML infrastructure at early-stage startups, including founding engineer at LEAP-ai and the first hire at PredictionIO. Based in Sunnyvale, he brings deep hands-on expertise in productionizing recommendation systems and extending core ML tooling, evidenced by significant open-source contributions to PredictionIO and PyTorch (notably Vulkan backend optimizations). He thrives in collaborative, fast-moving teams where refactoring and shifting data representations matter, and he has a track record of shipping low-level performance fixes as well as algorithmic improvements. Justin combines startup pragmatism with low-level systems chops from his Google and open-source background, often working on the subtle edge cases that make ML systems robust in production.
12 years of coding experience
Github Skills (19)
algorithm10
vulkan10
algorithms10
pytorch10
predictionio10
python10
machine-learning10
scala10
deeplearning-ai10
deep-learning10
gpu10
neural-network10
tensor10
c-language9
glsl9
Programming languages (6)
Github contributions (5)
apache/predictionio
Mar 2014 - Mar 2016
PredictionIO, a machine learning server for developers and ML engineers.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:62 commits, 5 PRs, 61 pushes in 2 years
Contributions summary:Justin primarily focused on refactoring output functionalities and modifying code related to item recommendation algorithms. They updated the code to pass sequences of items instead of identifiers, indicating a shift in data handling strategies. The user was involved in merging code changes, which suggests they were working within a collaborative environment. Further commits indicate work on feature-based item recommendation and related build processes.
pythondata-sciencemachine-learningmlopsbig-data
pytorch/pytorch
Feb 2023 - Mar 2024
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:43 reviews, 18 PRs, 1 push in 1 year 1 month
Contributions summary:Justin contributed to the PyTorch library by implementing and testing new functionalities for the Vulkan backend, specifically focusing on operators like `select`, `uniform`, `floor_divide`, `log` and `log_softmax`, and `conv1d`. Their work involved modifying existing code, adding new shader implementations, and creating tests. The user also addressed issues in the `cat` operator and enhanced the handling of zero-dimensional tensors within the Vulkan backend. This demonstrates their proficiency in extending PyTorch's functionality and optimizing it for specific hardware.
pythongpu-accelerationdeep-learninggpunumpy
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.