Pushkar Kolhe
Software Engineer at Snap Inc.
Greater Seattle Area United States
Summary
👤
Senior🎓
Top SchoolPushkar Kolhe is a software engineer with 16 years of experience building tools that help teams move faster, currently contributing at Snap in the Greater Seattle Area. He blends research rigor from PhD-level computer science work with hands-on engineering across machine learning and build systems, demonstrated by significant contributions to the ABAGAIL ML library and build/resolution improvements for the DART robotics toolkit. As Co-Founder and CTO of a startup and a former Autopilot engineer at Tesla, he pairs product-minded leadership with deep implementation skills in algorithms, CMake/build tooling, and model evaluation (e.g., adding confusion-matrix metrics and benchmark examples). Known for pragmatic open-source contributions that improve portability and developer workflows, he favors solutions that make complex systems easier for others to use and extend.
16 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at Georgia Tech
Stackoverflow
111reputation
1kreached
0answers
1question
Github Skills (19)
algorithm10
robotics10
optimizers10
machine-learning10
build-system10
cmake10
java10
javas10
optimisation10
neural-network10
optimization10
testing9
kinematics9
c-language9
dynamics9
Programming languages (4)
Github contributions (5)
pushkar/ABAGAIL
Mar 2013 - Feb 2017
The library contains a number of interconnected Java packages that implement machine learning and artificial intelligence algorithms. These are artificial intelligence algorithms implemented for the kind of people that like to implement algorithms themselves.
Role in this project:
ML Engineer
ML EngineerContributions:8 commits in 4 years
Contributions summary:Pushkar contributed significantly to the machine learning and artificial intelligence algorithms within the repository. They introduced a Confusion Matrix test metric, crucial for evaluating model performance. The user also added an Abalone test example, implementing various optimization algorithms like Randomized Hill Climbing, Simulated Annealing, and a Genetic Algorithm. Furthermore, modifications related to specific testing methods were made.
optimization-algorithmsartificialmachine-learningneural-networkartificial-intelligence-algorithms
dartsim/dart
Aug 2012 - Apr 2013
DART: Dynamic Animation and Robotics Toolkit
Role in this project:Back-end Developer / Build & Release Engineer
Back-end Developer / Build & Release EngineerContributions:45 commits in 8 months
Contributions summary:Pushkar primarily focused on build system improvements and ensuring proper dependencies. They added a CMake module for finding and using the Eigen3 library and modified the CMake build process to work better with pkg-config and to address potential errors in Windows builds related to Boost and GLUT. Furthermore, the user updated the build process for gtest to work directly from the Linux libgtest-dev package and created FindDart and FindDartExt CMake config files for managing external dependencies. These modifications streamlined the build process and improved the project's portability.
kinematicsroboticsanimationcomputer-animationsimulation
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.