Li Li
Machine Learning Engineer at Apple
Bellevue, Washington, United States
Summary
👤
Senior🎓
Top SchoolLi Li is a machine learning engineer in Bellevue with eight years of experience building large-scale ML compute and cloud-native systems, currently making ML compute "stress-free" at Apple. She holds a Ph.D. in Electrical Engineering (4.0 GPA) and began her career designing distributed ML compute at Microsoft, where she helped build Azure Batch AI from scratch and contributed to the Azure SDK for .NET. Her work spans production-grade MLOps—optimizing PyTorch BERT training with Horovod on AzureML clusters—and low-level statistical signal processing from her academic background. Colleagues rely on her rare combination of research rigor and pragmatic engineering to improve distributed training performance, observability, and deployment workflows. An active contributor to Microsoft open-source efforts, she pairs cloud-scale systems know-how with a deep foundation in sensing and mobile data processing.
8 years of coding experience
4 years of employment as a software developer
Bachelor's Degree, Communications Engineering, Bachelor's Degree, Communications Engineering at University of Electronic Science and TechnologyDoctor of Philosophy (Ph.D.), Electrical Engineering, 4.0/4.0, Doctor of Philosophy (Ph.D.), Electrical Engineering, 4.0/4.0 at Duke University
English, Chinese
Github Skills (19)
pytorch10
distributed-training10
azure-sdk10
net10
dotnet10
microsoft-azure10
azure-machine-learning10
asp-net10
sdk10
bert10
fine-tuning10
dotnet-core10
azure10
nlp10
csharp10
Programming languages (7)
Github contributions (5)
microsoft/AzureML-BERT
Dec 2018 - Feb 2019
End-to-End recipes for pre-training and fine-tuning BERT using Azure Machine Learning Service
Role in this project:
MLOps Engineer
MLOps EngineerContributions:9 commits, 6 pushes, 1 comment in 2 months
Contributions summary:Li primarily focuses on integrating and optimizing PyTorch-based BERT models within the Azure Machine Learning environment. Their contributions include modifying existing scripts to work seamlessly with AzureML, refactoring code for improved distributed training performance using Horovod, and updating notebook examples. They demonstrate expertise in configuring and running training jobs on Azure compute clusters and fine-tuning model hyperparameters.
bertfinetuningtrainingbert-fine-tuninglanguage-model
Azure/azure-sdk-for-net
Oct 2017 - Jun 2018
This repository is for active development of the Azure SDK for .NET. For consumers of the SDK we recommend visiting our public developer docs at https://learn.microsoft.com/dotnet/azure/ or our versioned developer docs at https://azure.github.io/azure-sdk-for-net.
Role in this project:Back-end Developer
Back-end DeveloperContributions:11 commits, 6 PRs, 8 comments in 8 months
Contributions summary:Li primarily contributed to the development and maintenance of the Azure BatchAI SDK for .NET. Their work included adding new SDKs, updating existing ones for GA releases, and incorporating changes to support new features such as job-level mounting, secret environment variables, and performance counters in Azure Application Insights. The commits also involved updating the SDK version and adding a PowerShell generation script. Further contributions include fixing build and playback issues and addressing comments in pull requests.
dotnetversionedazure-functionssdkconsumers
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.