Shoubhik Bhattacharya
Machine Learning Engineer at Facebook
Seattle, Washington, United States
Summary
👤
Senior🎓
Top SchoolShoubhik Bhattacharya is a Machine Learning Engineer in Seattle with 13 years of experience building large-scale distributed systems, ML data pipelines, and inference optimization platforms. He currently focuses on large-scale inference benchmarking and has contributed to the TVM deep learning compiler by implementing and optimizing quantized neural network operators. His background spans research and production roles at Amazon and Facebook, where he fused cloud-native engineering (AWS), C++/Python development, and software design to scale ML systems. He’s also worked on secure multi-party computation for federated learning, reflecting an interest in privacy-preserving ML beyond typical model engineering. Across early-career backend and cloud migrations to cutting-edge compiler work, he blends practical system design with open-source contributions that impact inference performance at scale.
13 years of coding experience
11 years of employment as a software developer
Master's degree, Computer Science, A, Master's degree, Computer Science, A at North Carolina State UniversityDeep Learning Nanodegree Foundation, Computer Science, A, Deep Learning Nanodegree Foundation, Computer Science, A at UdacityB.Tech, Computer, B.Tech, Computer at Uttar Pradesh Technical University
Github Skills (25)
tvm10
ruby-rails10
expert-system10
back-end-development10
mxnet10
machine-learning10
rails10
expert10
ruby10
relayer10
deeplearning-ai10
compiler-compiler10
deep-learning10
compiler10
tensor10
Programming languages (7)
Github contributions (5)
apache/tvm
Aug 2019 - Apr 2020
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
ML Engineer
ML EngineerContributions:13 commits, 25 PRs, 139 comments in 8 months
Contributions summary:Shoubhik primarily contributed to the implementation and optimization of quantized neural network (QNN) operators within the TVM compiler stack. Their work involved the creation and modification of operators such as quantize, dequantize, and the QNN variants of dense and convolution layers, specifically focusing on the MXNet frontend. They also added support for int32 to float32 dequantization. The contributions extended to creating and modifying unit tests for the newly added operators.
metalvulkancompilertensoropencl
expertiza/expertiza
Nov 2012 - Nov 2012
Expertiza is a web application through which students can submit and peer-review learning objects (articles, code, web sites, etc). The Expertiza project is supported by the National Science Foundation.
Role in this project:Back-end Developer
Back-end DeveloperContributions:24 commits in 16 days
Contributions summary:Shoubhik's commits primarily involve changes to the `SignUpSheetController.rb` and `AssignmentController.rb` files within the Expertiza application. These changes indicate a focus on modifying core features related to topic signup, assignment management, and due date handling. The user appears to be involved in enhancing the system's backend logic, including functionalities for managing topics, deadlines, and assignment settings. The commits also include test case fixes and updates to the schema file, suggesting a focus on code quality and database structure.
nationalpeer-reviewexpertiza-wikidjangoexpertiza
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.