Cesar Berrospi
Senior Research Scientist at IBM
Zurich, Zurich, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolCesar Berrospi is a senior research scientist based in Zurich with a strong mathematical engineering background from EPFL and over two decades of experience at IBM spanning research, analytics, and market analysis. He blends deep quantitative training with hands-on engineering, recently focusing on production-ready document processing for generative AI by improving XML parsers and backend pipelines for patent data. At IBM he has progressed from IT analyst to senior research roles, evidencing both technical depth and domain fluency in business analytics. Cesar’s open-source work on the docling project highlights a practical streak: converting complex, real-world formats (like USPTO patents) into consumable documents for downstream AI systems. He brings a pragmatic research mindset that turns formal methods into maintainable, test-covered software used in enterprise settings.
3 years of coding experience
5 years of employment as a software developer
Postgraduate Master, Mathematical Engineering, Postgraduate Master, Mathematical Engineering at Ecole polytechnique fédérale de LausannePostgraduate Master, Mathematical Engineering, Postgraduate Master, Mathematical Engineering at Ecole polytechniqueUPC Universitat Politècnica de Catalunya
Github Skills (11)
backend10
parsing10
python10
parse10
back-end-development10
xml-parsing10
document-processing10
html8
xlsx7
pdf7
docx7
Programming languages (3)
Github contributions (5)
docling-project/docling
Jul 2024 - Apr 2025
Get your documents ready for gen AI
Role in this project:
Back-end Developer
Back-end DeveloperContributions:37 reviews, 26 PRs, 31 pushes in 8 months
Contributions summary:Cesar focused on enhancing the back-end functionality of the document processing project. They primarily implemented and refactored XML backend parsers, particularly for USPTO patents, enabling the conversion of patent data into the project's document format. The user added support for new input formats and refined existing parsing logic. Furthermore, the user contributed to testing and documentation efforts, ensuring code quality and maintainability.
aiconvertdocument-parserdocument-parsingdocx
docling-project/docling-sdg
Mar 2025 - Apr 2025
A set of tools to create synthetically-generated data from documents
Contributions:5 reviews, 12 PRs, 12 pushes in 7 days
aillm-as-a-judgequestion-answeringsdg
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.