Leonidย Shamis
Staff Research Engineer at Meta
Mountain View, California, United States
Summary
๐คฉ
Rockstar๐
Top SchoolLeonid Shamis is a senior software and research engineer with 11+ years building scalable robotics, autonomy, and multi-modal AI systems from Google Brain to xAI in Mountain View. He has led infrastructure and inference optimization for large multi-modal models and owned agentic frameworks, RL training stacks, and actor/sampling infrastructure for production research. At Meta he shipped the Chameleon multi-modal model release and developed ALMA, a low-annotation LLM alignment approach, while earlier roles focused on motion planning, SLAM, and fleet orchestration for autonomous vehicles and robots. He contributes to high-performance open-source projectsโwork on yyjson added merge-patch and equality checks for a top C JSON library, and DevOps edits to Facebook Researchโs fairo show a knack for reproducible builds and dependency integration. Equally comfortable managing teams and shipping low-level C tests or Docker configs, he blends research rigor with production engineering discipline. Based in Mountain View, he pairs an MS in Computer Science with a track record of turning complex robotics and ML research into robust, deployable systems.
11 years of coding experience
13 years of employment as a software developer
Master of Science (M.S.) Computer Science, Master of Science (M.S.) Computer Science at University of California, Davis
Github Skills (14)
c1710
serialization10
deserialization10
json10
data-serialization10
docker10
c1110
dockers10
testing10
build-automation9
bash9
cross-platform9
cicd8
python6
Programming languages (15)
Github contributions (5)
facebookresearch/fairo
Aug 2021 - Aug 2022
A modular embodied agent architecture and platform for building embodied agents
Role in this project:
DevOps Engineer
DevOps EngineerContributions:70 reviews, 75 commits, 101 PRs in 1 year
Contributions summary:Leonid primarily focused on updating the repository's Docker configuration, specifically modifying the container name in the `local_install.sh` script. They also made changes related to migrating and integrating external libraries, like `fair-robotics/pyrobot-vision`. These modifications indicate involvement in build processes, environment setup, and integrating external dependencies related to embodied agents. The user also added a psutil logging feature.
agentagentsembodied-agentmodulararchitecture
ibireme/yyjson
Aug 2021 - Oct 2021
The fastest JSON library in C
Role in this project:Back-end Developer & QA Engineer
Back-end Developer & QA EngineerContributions:6 commits, 1 comment, 4 issues in 1 month
Contributions summary:Leonid primarily contributed to the development and testing of the yyjson library. They implemented the merge patch functionality, which involved adding new features and writing comprehensive test cases to ensure its correctness. Furthermore, the user added an equals function to compare JSON mut values. The user also improved code maintainability by fixing unused function warnings, and refactored by adding test cases and covering corner cases.
deserializationserializationfastestjson-libraryperformance
Find and Hire Top DevelopersWeโve 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.