Rakesh Vasudevan
Software Development Engineer at Amazon Web Services
California, United States
Summary
🤩
Rockstar🎓
Top SchoolRakesh Vasudevan is a Software Development Engineer with 10 years of experience building and productionizing ML and edge-inference systems, currently working on AWS Edge ML device SDKs and stacks. He has deep hands-on experience in model serving and framework integration, contributing to notable open-source projects like AWS Multi Model Server and the Deep Java Library where he implemented malicious-URL detection, model trainers, and Gluon character CNN examples. Rakesh pairs ML engineering with strong QA/test automation skills—his contributions to NumPy and Apache MXNet improved test stability and cross-platform correctness. His background spans silicon-to-software roles at MIPS and an Android Dalvik internship at Intel, giving him practical insight into low-level and embedded constraints. Based in California, he blends systems-level problem solving with ML inference optimization, often surfacing non-obvious platform issues such as GPU build test failures and backend-specific numerical bugs.
10 years of coding experience
4 years of employment as a software developer
Anna University, ChennaiMaster of Science (M.S.), Master of Science (M.S.) at Portland State UniversityPSBB
Github Skills (33)
pytorch10
python10
testing10
mxnet10
machine-learning10
inference10
java10
numpy10
mask-rcnn10
javas10
deep-learning10
tensorflow10
trainings10
ai10
neural-network10
Programming languages (7)
Github contributions (5)
awslabs/multi-model-server
Apr 2018 - Apr 2019
Multi Model Server is a tool for serving neural net models for inference
Role in this project:
ML Engineer
ML EngineerContributions:108 commits, 88 PRs, 35 pushes in 1 year
Contributions summary:Rakesh primarily contributed to the development and enhancement of model serving capabilities within the repository. They focused on improving the handling of different MXNet and ONNX models, including the integration of Gluon models. Their work involved the modification of service files, improvement of error propagation, and updates to documentation. The user also added a Gluon character CNN example, demonstrating the use of Gluon models for character-level CNN tasks.
pytorchmxnetservingdeep-learninginference
deepjavalibrary/djl-demo
Nov 2019 - Nov 2019
Demo applications showcasing DJL
Role in this project:ML Engineer
ML EngineerContributions:18 commits, 1 PR, 1 comment in 14 days
Contributions summary:Rakesh primarily worked on a project focused on malicious URL detection using deep learning. They were involved in the design and implementation of a model trainer and the core malicious URL model. Their work included defining and loading the model, implementing inference logic, and integrating the model into a request handler for real-time URL analysis. They also optimized the training process and updated training parameters.
demo-applicationsmxnetdeep-learningaidjl
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.