Stefano Fioravanzo
Freelance at Product Heroes
Trentino – Alto Adige/Südtirol, Italy
Summary
👤
Senior🎓
Top SchoolStefano Fioravanzo is a product-focused AI and MLOps leader with 11 years of experience building cloud-native ML platforms and developer tooling from code to production. Based in Trentino, Italy, he blends hands-on engineering (notably converting notebooks into Kubeflow Pipelines via the Kale project) with product strategy, community building, and technical storytelling. He has led AI product strategy at Canonical and HPE/Arrikto, driving multi-tenant, GPU-accelerated, and SaaS MLOps solutions while also improving UX in Kubeflow Pipelines. As Kubeflow ML Experience WG Lead and Chair of the Kubeflow Outreach Committee at CNCF, he pairs open-source stewardship with pragmatic delivery. Known for refactoring both front-end and back-end systems, he favors developer-centric, Jupyter-first experiences that shorten the path from experiment to production. Currently freelancing, he advises and ships short-term AI and cloud-native initiatives with a slightly opinionated, user-first approach.
10 years of coding experience
9 years of employment as a software developer
Master of Science (MS), Computer Science - Data Science, Master of Science (MS), Computer Science - Data Science at Università di Trento
Italian, English, German
Stackoverflow
1reputation
1kreached
0answers
1question
Github Skills (24)
data-pipelines10
back-end-development10
python10
kubeflow10
front-end-development10
pipe10
data-pipeline10
react10
dependency-management9
dockers9
typescript9
docker9
typescript-types9
typescripts9
backend8
Programming languages (12)
Github contributions (5)
kubeflow-kale/kale
Jan 2019 - May 2021
Kubeflow’s superfood for Data Scientists
Role in this project:
Back-end Developer & MLOps Engineer
Back-end Developer & MLOps EngineerContributions:7 releases, 50 reviews, 589 commits in 2 years 4 months
Contributions summary:Stefano was instrumental in developing the back-end functionality for converting notebooks into KFP pipelines. They implemented a converter module, contributing significantly to the project's core functionality by enabling the creation of pipelines from notebook-based code. The user also integrated and refactored code to improve data handling by implementing object serialization with `dill` and contributed to the infrastructure side of the project, by adding support to deploy the generated python code to KFP, including the usage of volume mounts.
scientistsdata-sciencejupyter-notebookmachine-learningdata-scientists
kubeflow/pipelines
Dec 2020 - Mar 2021
Machine Learning Pipelines for Kubeflow
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:43 reviews, 8 commits, 8 PRs in 2 months
Contributions summary:Stefano primarily worked on the frontend components of the Kubeflow Pipelines project, focusing on UI/UX improvements. Their contributions involved restructuring pages, adding new components (like the "Runs" sidenav item and a graph reduction switch), and integrating them into the application. They also refactored existing components and made updates to the related snapshots. The user also made changes to the backend to include features that supported the changes to the frontend, such as macro expansion for pipeline parameters.
pipelinetektondata-sciencemachine-learningmlops
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.