Sukhad Joshi
Software Engineer at Arista Networks
Raleigh, North Carolina, United States
Summary
👤
Senior🎓
Top SchoolSukhad Joshi is a software engineer with nine years of experience blending ML research and production engineering, currently building at Arista Networks after completing an MS in Computer Science at NC State. He has practical experience in privacy-preserving ML from contributions to the well-known OpenMined/PySyft project, where he implemented SMPC support for core layers and improved PyTorch serialization behaviors. His background spans research roles—automating data collection and predictive modeling for software health—and hands-on data engineering, where he led GCP migrations and cut dashboard render times by 70%. Sukhad is skilled at speeding CI/CD and containerized workflows for HPC workloads and enjoys tackling the intersection of scalable data pipelines and trustworthy ML. Notably, he combines open-source collaboration with mentorship and a proven track record of making complex systems more efficient and auditable.
9 years of coding experience
3 years of employment as a software developer
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at North Carolina State UniversityBachelor of Engineering, Computer Science, Bachelor of Engineering, Computer Science at Pune Institute of Computer Technology
Github Skills (12)
secure-computation10
pytorch10
python10
federated-learning10
cryptography10
test-automation9
serialization9
deeplearning-ai9
data-serialization9
deep-learning9
debug8
debugging8
Programming languages (4)
Github contributions (5)
OpenMined/PySyft
Oct 2019 - Apr 2020
Perform data science on data that remains in someone else's server
Role in this project:
ML Engineer
ML EngineerContributions:5 commits, 14 PRs, 56 comments in 5 months
Contributions summary:Sukhad contributed to the development of PySyft, a library for privacy-preserving machine learning. Their work included implementing an `ObjectNotFoundError` exception and fixing typos within the code base. A significant portion of their effort involved adding serialization support for `grad_fn` in `AutogradTensor` and implementing dropout support for secure multi-party computation (SMPC), demonstrating expertise in PyTorch and related privacy-focused techniques. They also removed implicit inplace `fix_precision` for `PointerTensor`.
pytorchcryptographyacquiringpythonscience
SE-Fall23-Group9/ScheduleBot
Oct 2023 - Oct 2023
ScheduleBot is a Discord bot that helps your everyday life by helping you schedule things out. Talk with the bot privately to find a hole in your schedule for that thing you want to do. Or automatically find time and organize a meeting with many people by commanding the bot in your server. ScheduleBot is built on the python programming language, us
Contributions:1 release, 4 reviews, 8 PRs in 11 days
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.