Krishna Jatavallabhula
Visiting Scholar at Toyota Research Institute
Berkeley, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKrishna Jatavallabhula is an incoming assistant professor at Johns Hopkins University and a visiting scholar at Toyota Research Institute, bringing 11 years of deep robotics and AI experience across industry and academia. His work spans the full robotics stack—from differentiable SLAM and NeRF implementations to sensing, control interfaces, state estimation, and high-level robot learning—delivered at institutions like Meta, MIT CSAIL, Mila, and NVIDIA. He has been recognized by Nvidia and Google fellowships and was inducted into the RSS Pioneers 2020 cohort, reflecting influential contributions to robotics science. Krishna is an active open-source contributor—his work on gradslam added core dataset and geometry utilities crucial for dense, differentiable SLAM in PyTorch and he’s extended NeRF tooling for reproducible research. Based in Berkeley, he blends hands-on engineering with rigorous research, frequently moving ideas from prototype code to peer-reviewed impact. A not-obvious strength is his cross-layer fluency: he routinely connects low-level geometric data handling to high-level world models, enabling practical advances in embodied perception.
11 years of coding experience
4 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and ScienceDoctor of Philosophy - PhD, Robotics, Computer Vision, Deep Learning, Doctor of Philosophy - PhD, Robotics, Computer Vision, Deep Learning at Université de MontréalMaster’s Degree, Computer Science, Master’s Degree, Computer Science at International Institute of Information Technology Hyderabad (IIITH)
Telugu, Hindi, English
Github Skills (17)
3d-reconstruction10
pytorch10
serf10
geometry10
python10
machine-learning10
deep-learning10
3d10
slam10
geom10
data-structure9
configuration-management9
data-structures9
computer-vision9
tensorboard8
Programming languages (12)
Github contributions (5)
krrish94/nerf-pytorch
Mar 2020 - Apr 2020
A PyTorch re-implementation of Neural Radiance Fields
Role in this project:
ML Engineer
ML EngineerContributions:57 commits, 5 PRs, 37 pushes in 20 days
Contributions summary:Krishna's commits primarily involve the implementation and extension of a PyTorch-based neural radiance field (NeRF) model. They have added core functionalities such as config file parsing, dataset caching for the Lego dataset, and multi-head and replicate NeRF model architectures. The contributions also include integrating a progress bar (tqdm) and expanding the logging capabilities, all centered around enhancing NeRF model training and validation.
pytorchre-implementationneural-radiance-fieldscomputer-graphicscomputer-vision
gradslam/gradslam
Feb 2020 - May 2021
gradslam is an open source differentiable dense SLAM library for PyTorch
Role in this project:ML Engineer
ML EngineerContributions:7 reviews, 46 commits, 15 PRs in 1 year 3 months
Contributions summary:Krishna primarily contributed to the `gradslam` library by adding and modifying code related to datasets and geometry functions. They added a new `ScannetDataset` class for loading and preprocessing data. Furthermore, the user implemented several geometry utility functions, including point cloud transformations, quaternion conversions, and projection utilities, which are core to SLAM algorithms. These changes suggest a focus on the core data handling and geometric aspects of the SLAM pipeline.
pytorchroboticsdifferentiabledeep-learning3d-reconstruction
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.