Kevin Doherty
Research Engineer at Boston Dynamics
Cambridge, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolKevin Doherty is a research engineer at Boston Dynamics with 11 years of experience building perception and state-estimation systems for real-world robots. He completed a PhD in the MIT/WHOI Joint Program and spent six years in CSAIL’s Marine Robotics Group developing learning-backed navigation and long-term autonomy methods. Kevin combines principled probabilistic inference—demonstrated by contributions to the widely used GTSAM library improving numerical stability of discrete factor graphs—with practical deployment on complex platforms like Atlas. His work focuses on enabling robots to operate robustly over long horizons through active, lifelong learning and robust representations. Based in Cambridge, he brings a rare blend of academic rigor and hands-on system integration across aerial, underwater, and humanoid robots.
11 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Aeronautics and Astronautics & Ocean Engineering, Doctor of Philosophy - PhD, Aeronautics and Astronautics & Ocean Engineering at Massachusetts Institute of TechnologyBachelor of Engineering (BEng), Electrical and Electronics Engineering, Bachelor of Engineering (BEng), Electrical and Electronics Engineering at Stevens Institute of TechnologyHigh School Diploma, High School Diploma at Jackson Memorial High School
English
Stackoverflow
1reputation
66reached
0answers
1question
Github Skills (18)
algorithm10
algorithms10
c-language10
testing10
graph10
gtsam10
data-structure10
factors10
numerical-methods10
bayesian-network10
data-structures10
cprogramming-language10
robotics9
python9
machine-vision9
Programming languages (10)
Github contributions (5)
borglab/gtsam
Mar 2018 - Jan 2025
GTSAM is a library of C++ classes that implement smoothing and mapping (SAM) in robotics and vision, using factor graphs and Bayes networks as the underlying computing paradigm rather than sparse matrices.
Role in this project:
Back-end Developer / Algorithm Implementer
Back-end Developer / Algorithm ImplementerContributions:5 reviews, 2 PRs, 1 push in 6 years 11 months
Contributions summary:Kevin focused on improving the numerical stability and accuracy of the GTSAM library's discrete factor graph implementations. Their contributions involved adding normalization techniques to prevent underflow in max-product and sum-product algorithms, essential for robust probabilistic inference. They also introduced and refined test cases designed to specifically identify and address potential underflow issues within these algorithms, ensuring their reliability. Furthermore, the user refactored the existing code to simplify the normalization methods.
smoothingsparsec-plus-plushierarchicalpattern-recognition
MarineRoboticsGroup/mac
Mar 2022 - Jul 2022
Maximizing algebraic connectivity for graph sparsification
Contributions:40 commits, 8 PRs, 49 pushes in 4 months
algebraic-connectivitygraph-slamgraph-sparsificationpose-graph-optimizationspectral-graph-theory
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.