Nikhil Keetha
Visiting Researcher at The AirLab at Carnegie Mellon University
Pittsburgh, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolNikhil Keetha is a robotics and computer vision researcher with six years of experience focused on enabling machines to see and reason about the world. Currently a Visiting Researcher at Meta XR Maps and a Graduate Research Assistant at CMU’s AirLab, he leads perception work in localization, reconstruction, and SLAM under advisors Sebastian Scherer and Deva Ramanan. His background spans research internships at Mila, QUT, and PURE EV, where he explored multi-view consistent representations, semantic place recognition, and applied ML to IoT battery data. A PhD/MS Robotics student at Carnegie Mellon and IIT ISM alumnus, he combines strong theoretical grounding with practical systems experience across academia and industry. Notably, he has developed temporally evolving graph-based object representations for class-agnostic object identification and contributed to GradSLAM-related research bridging NeRFs and continual learning for SLAM.
6 years of coding experience
1 year of employment as a software developer
Bachelor of Technology with Distinction Engineering Physics, Bachelor of Technology with Distinction Engineering Physics at Indian Institute of Technology (Indian School of Mines), DhanbadClass XII, Class XII at FIITJEE Dilsukhnagar HyderabadDoctor of Philosophy - PhD Robotics, Doctor of Philosophy - PhD Robotics at Carnegie Mellon UniversityClass X, Class X at Slate-the-school
Github Skills (73)
depth-estimation10
robotics10
neural-rendering10
localization10
ai10
computer-vision10
3d-reconstruction10
twitter-sentiment-analysis10
image-retrieval10
calibration10
slam10
contrastive-learning10
sentiment-analysis9
resnet9
image-matching9
Programming languages (6)
Github contributions (5)
facebookresearch/map-anything
Sep 2025 - Jul 2026
MapAnything: Universal Feed-Forward Metric 3D Reconstruction
Contributions:1 release, 2 reviews, 5 PRs in 10 months
3d-reconstructionaicalibrationdepth-completiondepth-estimation
AnyLoc/AnyLoc
Aug 2023 - Nov 2025
AnyLoc: Universal Visual Place Recognition (RA-L 2023)
Contributions:2 PRs, 2 pushes, 1 branch in 2 years 3 months
deep-learningimage-retrievallocalizationplace-recognitionrobotics
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.