Shagun Sodhani
Staff Research Scientist at Google DeepMind
India
Summary
🤩
Rockstar🎓
Top SchoolShagun Sodhani is a Staff Research Scientist with 13 years of experience bridging machine learning research and production-grade software engineering, currently at Google DeepMind after leading AI research efforts at Meta. He has co-led initiatives on autonomous research agents, trained foundation models for adaptive neural interfaces, and contributed to 10+ top-tier publications across representation learning, lifelong learning, and RL. A seasoned open-source contributor, he fixed core bugs and built an Ax hyperparameter optimization plugin for Facebook Research’s Hydra, improving experiment configuration and reproducibility. Trained at MILA under Yoshua Bengio and with industry stints at Adobe and Microsoft Research, he pairs deep academic grounding with practical product and tooling impact. Known for mentoring teams and shipping ML toolkits, he often focuses on making complex research workflows robust and automatable.
12 years of coding experience
13 years of employment as a software developer
Masters Artificial Intelligence, Masters Artificial Intelligence at Université de MontréalIndian Institute of Technology RoorkeeHigh School High School/Secondary Diplomas and Certificates, High School High School/Secondary Diplomas and Certificates at Somerville School
English, Hindi
Stackoverflow
3,605reputation
807kreached
35answers
16questions
machine-learning
top-5%
apache-spark
top-5%
python
top-5%
pytorch
top-5%
gpu
top-5%
Github Skills (14)
configuration-management10
python10
pytorch9
pytest9
testing9
gpu9
machine-learning9
apache-spark9
softmax6
logistic-regression6
makefile6
tuples6
numpy6
cygwin6
Programming languages (14)
Github contributions (5)
facebookresearch/hydra
Jan 2020 - Feb 2021
Hydra is a framework for elegantly configuring complex applications
Role in this project:
Back-end Developer
Back-end DeveloperContributions:128 reviews, 78 commits, 129 PRs in 1 year 1 month
Contributions summary:Shagun primarily contributed to the `facebookresearch/hydra` repository by fixing bugs related to the `utils.instantiate()` function, ensuring it correctly handles interpolated values. They also developed a plugin to integrate the Ax hyperparameter optimization platform with Hydra, streamlining the experimentation process. Further, they implemented improvements to testing and logging of the Ax plugin.
pytorch-lightningcomplex-applicationsframeworkhydra
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.