Daniele Trifirò
Principal Software Engineer at Red Hat
Milan, Lombardy, Italy
Summary
🤩
Rockstar🎓
Top SchoolDaniele Trifirò is a Principal Software Engineer based in Milan with 7 years of experience building cloud-native, ML and DevOps systems using Python, Go, Rust and C/C++. He combines a physics background from the University of Pisa with hands-on expertise in Kubernetes, Docker, Terraform, Azure and MLOps to bridge model development and production-grade deployment. At Red Hat he has driven OpenShift AI and inference engineering work, and previously contributed to DVC and the pure-Python Git implementation dulwich, adding TOML params support and improved untracked-file handling respectively. He also improved high-throughput LLM build and CI workflows for the vLLM project, showing a knack for robust build systems and reproducible ML stacks. Colleagues rely on him for pragmatic, low-level fixes and CI/build optimizations that noticeably reduce developer friction.
7 years of coding experience
11 years of employment as a software developer
Bachelor's degree, Physics, Bachelor's degree, Physics at Università di Pisa
Github Skills (25)
git10
rep10
docker10
pytest10
python10
machine-learning10
build-system10
dockers10
cicd10
toml10
data-versioning10
repr10
version-control10
versioning10
dulwich10
Programming languages (12)
Github contributions (5)
iterative/dvc
Jul 2020 - Jan 2023
🦉 Data Versioning and ML Experiments
Role in this project:
Back-end Developer
Back-end DeveloperContributions:107 reviews, 43 commits, 64 PRs in 2 years 6 months
Contributions summary:Daniele primarily contributed to the data versioning and machine learning experiment tracking project, DVC. Their work focused on adding TOML support for the ParamsDependency feature, including implementing the ability to read and display parameter configurations from TOML files. This involved modifying core DVC files related to dependency management and parameter display. Additionally, the user made contributions related to fixing and improving test cases.
data-version-controlml-modelspythonversion-controldata-science
vllm-project/vllm
Apr 2024 - Apr 2025
A high-throughput and memory-efficient inference and serving engine for LLMs
Role in this project:DevOps Engineer & Build/Release Engineer
DevOps Engineer & Build/Release EngineerContributions:109 reviews, 32 PRs, 201 comments in 11 months
Contributions summary:Daniele's commits primarily focused on improving and maintaining the build and CI/CD processes for the vLLM project. They made significant changes to Dockerfiles, enhancing the CPU build process and ensuring compatibility with podman. The user also implemented improvements to the build system, including enabling ccache/sccache for HIP builds and integrating setuptools-scm for version management. Furthermore, they addressed issues related to the build process, such as fixing the pre-compiled wheel installation and adding checks for dirty repositories.
amdcudadeepseekgpthpu
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.