Bowen Chen
Summary
🤩
RockstarBowen Chen is a software engineer with eight years of experience specializing in ML systems, inference optimization, and AI framework development. He’s delivered production LLM inference at AWS—optimizing latency, throughput, sharding strategies, and quantization for accelerators like Inferentia and Trainium—while contributing core numerical and module-level work to the OneFlow deep learning framework. Bowen combines research rigor from a Carnegie Mellon MCDS and multiple academic research assistantships with hands-on engineering across startups and hyperscalers, from AI framework internships to building disaggregated inference solutions. Notably, his open-source commits to OneFlow focus on convolutional backward modules, RNG integration, and dropout/bernoulli implementations—signal work that underpins model correctness and performance. Now at Fireworks AI, he continues to bridge low-level algorithmic detail and large-scale system design to push efficient, production-grade ML deployments.
8 years of coding experience
Github Skills (9)
neural-network10
cuda10
machine-learning10
deep-learning10
deep-neural-networks10
python10
convolutional-neural-networks9
randomization9
test-automation8
Programming languages (10)
Github contributions (5)
Oneflow-Inc/oneflow
Jun 2021 - Feb 2022
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.
Role in this project:
ML Engineer
ML EngineerContributions:80 reviews, 105 commits, 43 PRs in 8 months
Contributions summary:Bowen's commits primarily focus on the development and testing of modules within the OneFlow deep learning framework. Their work includes adding backward modules for convolutional layers and integrating new random number generator functionalities, which are vital components for the framework's numerical computation and training processes. Additionally, they've refactored dropout modules and implemented bernoulli functions. The user's contributions strongly align with the core functionality and underlying mechanisms of deep learning models, indicating a focus on the framework's mathematical and algorithmic aspects.
caffe2tensorflowframework-learningchainerdeep-learning-framework
VertexC/dl-infer-perf
Feb 2021 - Jun 2021
Contributions:96 commits, 39 pushes, 6 comments in 4 months
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.