Mor Zusman
Principal Full Stack Engineer at Boomi
Tel-Aviv, Tel-Aviv District, Israel
Summary
🤩
Rockstar🎓
Top SchoolMor Zusman is a Principal Full Stack Engineer based in Tel-Aviv with a B.Sc. in Computer Science and a decade of hands-on experience across image processing, Python back-end systems, and full-stack development. He has progressed from implementing image algorithms at Rafael to leading a full-stack team at Connecteam and now driving engineering at Boomi after Rivery's acquisition. Mor brings strong backend expertise (Python, Tornado), frontend experience (React, Angular, TypeScript), and production cloud familiarity across AWS and GCP for large-scale SaaS and IoT systems. He has led onboarding and feature ownership efforts, mentored teams, and managed operational concerns like monitoring and GDPR integrations. An active contributor to high-performance ML infra, he fixed nuanced model-execution issues in the popular vllm inference engine, demonstrating attention to edge-case stability and numerical subtleties. Colleagues rely on him for pragmatic, full-stack solutions that bridge algorithms, data pipelines, and scalable production services.
10 years of coding experience
7 years of employment as a software developer
Computer Science, Computer Science at University of Haifa
Github Skills (8)
transformer10
pytorch10
inference10
llm10
serve9
lm9
cuda9
p88
Programming languages (3)
Github contributions (5)
vllm-project/vllm
Apr 2024 - Feb 2025
A high-throughput and memory-efficient inference and serving engine for LLMs
Role in this project:
ML Engineer
ML EngineerContributions:118 reviews, 13 PRs, 158 comments in 10 months
Contributions summary:Mor primarily contributed to the VLLM project, a high-throughput and memory-efficient inference engine for LLMs, by addressing issues related to model execution and inference. Their work included fixing a bug related to dummy weights for FP8 data types, which involved modifying weight initialization to accommodate the limitations of torch.uniform_. They also contributed to supporting the Jamba model by implementing the model support with necessary changes in the codebase. The user also addressed issues in model execution pipeline like multi-step and varlen prefill.
amdcudadeepseekgpthpu
mzusman/vllm
Apr 2024 - Mar 2025
A high-throughput and memory-efficient inference and serving engine for LLMs
Contributions:191 pushes, 18 branches, 1 comment in 10 months
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.