Max De Bayser is a Senior Research Software Engineer with 14 years of experience building foundational software—from OS-level components and C++ frameworks to cloud orchestration and AI inference systems. At IBM he bridges prototypes to products, contributing to watsonx.ai and high-throughput LLM serving work such as the popular vLLM project where he fixed critical multiprocessing and tool-use issues. He excels at systems thinking, designing for performance, scalability and developer productivity across the full stack, and has hands-on MLOps experience scaling LLM training data pipelines with Ray. Earlier research work at CPTI/PUC-Rio gave him deep exposure to embedded and signal-processing software, reinforcing a pragmatic, cross-disciplinary approach. Colleagues value his global collaboration style and focus on building reusable foundations that enable others to deliver.
A high-throughput and memory-efficient inference and serving engine for LLMs
Role in this project:
Back-end & MLOps Engineer
Contributions:104 reviews, 29 PRs, 173 comments in 10 months
Contributions summary:Max primarily focused on bug fixes and feature enhancements within the VLLM project, which focuses on serving and inference of LLMs. Their contributions included resolving issues related to server crashes, especially in the context of multiprocessing and empty prompts. They also added support for new tool use features in the Llama 3 models, and implemented tool call fixes. This suggests involvement in the core engine functionality and API interactions.
Contributions:195 commits, 60 pushes in 4 years 2 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.
Request Free Trial
Max De Bayser - Senior Research Software Engineer at IBM