Carolyn Wang
Software Engineer at Anyscale
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolCarolyn Wang is a software engineer based in San Francisco with eight years of experience building cloud-native infrastructure and ML-focused developer tools. She has driven persistent compute cluster infrastructure for large model training at AWS and now contributes to Anyscale, blending systems engineering with MLOps best practices. Her open-source work on Kubeflow Pipelines—adding SageMaker components for HPO, training, and deployment plus container/CI improvements—underscores a focus on making ML workflows reproducible and production-ready. A UC Berkeley EECS graduate with a statistics minor, she pairs rigorous CS fundamentals with practical performance wins (e.g., order-of-magnitude speedups in prior MongoDB work). Colleagues know her for creating accessible, high-impact solutions and for learning by shipping incremental, well-tested improvements.
8 years of coding experience
5 years of employment as a software developer
Bachelor of Science - BS Electrical Engineering & Computer Science Minor in Statistics, Bachelor of Science - BS Electrical Engineering & Computer Science Minor in Statistics at University of California, BerkeleyThomas Jefferson High School for Science and Technology
Chinese, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (15)
docker10
kubeflow-pipelines10
data-pipeline10
data-pipelines10
python10
dockers10
pipe10
pipeline10
kubernetes-pods9
mlops9
kubernetes9
yaml8
cicd8
data-science7
data-engineering6
Programming languages (4)
Github contributions (5)
kubeflow/pipelines
Jul 2019 - Aug 2019
Machine Learning Pipelines for Kubeflow
Role in this project:
MLOps Engineer
MLOps EngineerContributions:5 commits, 5 PRs, 25 comments in 19 days
Contributions summary:Carolyn contributed significantly to the development and enhancement of SageMaker components within the Kubeflow Pipelines repository. Their work involved adding and updating components for Hyperparameter Tuning (HPO), training, and batch transform, along with integrating these components into sample pipelines, specifically an MNIST classification pipeline. The user also added components for workteam and Ground Truth, alongside a demo pipeline, showcasing a focus on machine learning operations and model deployment within the Kubeflow environment. Numerous Dockerfile updates and image versioning adjustments are present, indicating expertise in containerization and CI/CD within a machine learning workflow.
pipelinetektondata-sciencemachine-learningmlops
carolynwang/carolynwang.github.io
May 2021 - Sep 2024
Contributions:117 pushes, 3 branches in 3 years 4 months
cssreactnextjsshowcase
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.