Prabhat Nagarajan
Edmonton, Alberta, Canada
Summary
🤩
Rockstar🎓
Top SchoolPrabhat Nagarajan is a reinforcement learning researcher and software engineer with 11 years of experience spanning industry labs and academic research, currently pursuing a PhD in Statistical Machine Learning at the University of Alberta. He has applied RL to real-world problems during internships at Microsoft Research and SonyAI and built production-focused RL systems as an engineer at Preferred Networks in Tokyo. Prabhat contributes to widely used open-source RL toolkits (ChainerRL, Chainer, PFRL), notably improving replay buffer modularity and prioritized n-step transitions—work that underpins stable deep RL training. With dual degrees in Computer Science and Mathematics from UT Austin, he bridges rigorous theory and practical systems, and has taught ML foundations to industry partners through Amii. Colleagues describe him as a careful engineer who combines documentation rigor with low-level algorithmic improvements, making complex RL components easier to use and reuse.
10 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD, Computing Science - Statistical Machine Learning, Doctor of Philosophy - PhD, Computing Science - Statistical Machine Learning at University of AlbertaMaster of Science (MS), Computer Science, Master of Science (MS), Computer Science at The University of Texas at Austin
English, Tamil, Japanese
Github Skills (15)
neural-network10
deep-reinforcement-learning10
pytorch10
machine-learning10
deep-learning10
python10
reinforcement-learning10
documentation10
chainer10
ml-deployment9
continuous-deployment9
algorithms9
data-structures8
testing8
data-structure8
Programming languages (7)
Github contributions (5)
pfnet/pfrl
Jun 2020 - Nov 2021
PFRL: a PyTorch-based deep reinforcement learning library
Role in this project:
ML Engineer
ML EngineerContributions:24 reviews, 118 commits, 43 PRs in 1 year 4 months
Contributions summary:Prabhat primarily updated and modified the `pretrained_models.py` file, which suggests a focus on managing and utilizing pre-trained models within the PFRL library. These updates involved adding environment variables and modifying file paths for model downloads, indicating a contribution to model management and deployment. The user's work also included the addition of A3C tests, demonstrating the implementation and verification of deep reinforcement learning algorithms.
pytorchactor-criticdeep-learningreinforcement-learningdeep-reinforcement-learning
chainer/chainerrl
Sep 2018 - Apr 2021
ChainerRL is a deep reinforcement learning library built on top of Chainer.
Role in this project:Backend Developer
Backend DeveloperContributions:2 reviews, 584 commits, 148 PRs in 2 years 7 months
Contributions summary:Prabhat's commits primarily focused on refactoring and enhancing the replay buffer functionality, a core component of the deep reinforcement learning library. They implemented and extended features for n-step transitions and prioritized replay, and improved the modularity of the buffer for easier use. Their work involved modifying the behavior of the replay buffer to facilitate complex deep learning models.
actor-criticdeep-learningreinforcement-learningdeep-reinforcement-learningchainer
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.