Manu Seth
Software Development Engineer at Amazon Web Services
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolManu Seth is a Software Development Engineer with nine years of experience building and hardening deep learning infrastructure at AWS, based in Mountain View. He combines a strong academic foundation in machine learning (UCSD MS) and electrical engineering (IIT Kanpur) with hands-on expertise in GPU-accelerated algorithms and test automation for major open-source frameworks like Apache MXNet. At AWS he focuses on deep learning frameworks, and his prior work includes implementing GPU samplers and a Box–Cox operator for MXNet during a Core ML internship, as well as stabilizing flaky GPU operator tests and adding targeted unit tests to improve reliability. Early roles in analytics and systems design sharpened his ability to translate statistical models and low-level engineering into production-impacting features. Known for taking full ownership of projects, he bridges research-grade algorithms and production quality engineering to reduce latency and increase robustness.
9 years of coding experience
University of California San DiegoIndian Institute of Technology Kanpur
English, Hindi
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (7)
unit-testing10
mxnet10
python10
test-automation10
testing10
deep-learning9
gpu8
Programming languages (4)
Github contributions (5)
apache/mxnet
Aug 2018 - Dec 2021
Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
Role in this project:
QA Engineer / Test Automation Engineer
QA Engineer / Test Automation EngineerContributions:153 reviews, 75 commits, 160 PRs in 3 years 5 months
Contributions summary:Manu primarily focused on improving the quality and reliability of the MXNet deep learning framework through test-related commits. Their work included fixing flaky test issues in the GPU operator tests (convolution and depthwise convolution), preventing tests from failing due to incorrect seed usage, and adding new unit tests for operators like MKLDNN FullyConnected and Softmax. The user also addressed large-input issues related to softmax operator and enabled previously flaky tests. The user’s contributions were primarily focused on strengthening the testing infrastructure.
pythonschedulerdataflowmutationdata-science
mseth10/incubator-mxnet
Jul 2018 - Dec 2021
Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
Contributions:4 PRs, 596 pushes, 149 branches in 3 years 5 months
pythonschedulerdataflowmutationorchestration
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.