Edouard Belval
Senior Research Engineer at Amazon Web Services (AWS)
Greater Montreal Metropolitan Area Canada
Summary
🤩
Rockstar🎓
Top SchoolEdouard Belval is a Senior Research Engineer at AWS with a decade of experience building computer vision and document intelligence systems, currently working on Bedrock data automation and Amazon Quick/Quick Desktop. He progressed through multiple roles in Rekognition and Textract, combining research-grade algorithms with production engineering to ship robust OCR and document-processing features. A hands-on full-stack contributor, his open-source work includes pdf2image and a widely used synthetic TextRecognitionDataGenerator that supports multi-language, distorted and handwritten-style text for training OCR models. He frequently improves usability through clear documentation and example notebooks, as seen in contributions to Amazon Textract sample projects. Based in Greater Montreal, he pairs a Mila master's background with practical system-building experience across startups and enterprise teams, and often bridges model research and scalable backend implementations.
10 years of coding experience
8 years of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at Mila - Quebec Artificial Intelligence InstituteBachelor's degree, Computer Software Engineering, Bachelor's degree, Computer Software Engineering at École de technologie supérieureHigh School, General Studies, High School, General Studies at Collège Saint-HilaireTechnique informatique de gestion, Computer Science, Technique informatique de gestion, Computer Science at Cégep de Saint-Hyacinthe
English, French
Github Skills (23)
text-generation10
python10
args10
image10
technical-writing10
pi10
argparse10
argument-parsing10
image-manipulation10
pdf-conversion10
documentation10
image-processing9
testing9
fileio9
file-access9
Programming languages (12)
Github contributions (5)
Belval/pdf2image
May 2017 - Dec 2022
A python module that wraps the pdftoppm utility to convert PDF to PIL Image object
Role in this project:
Back-end Developer
Back-end DeveloperContributions:18 releases, 12 reviews, 170 commits in 5 years 8 months
Contributions summary:Edouard primarily contributed to the core functionality of the `pdf2image` module, which involves converting PDF files to images. Their commits include implementing the main logic for converting PDFs from both file paths and byte streams using the `pdftoppm` utility. They also fixed import errors and added testing infrastructure, improving the module's reliability. Furthermore, they added the option to output grayscale images and features like size options.
pythonpdfconvert-pdfpopplerwraps
Belval/TextRecognitionDataGenerator
Jul 2017 - Aug 2022
A synthetic data generator for text recognition
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:8 releases, 22 reviews, 122 commits in 5 years 2 months
Contributions summary:Edouard primarily contributed to the development of a synthetic data generator for text recognition, evidenced by the initial commit introducing the core functionality. The subsequent commits demonstrate enhancements to the generator, including fixing image resizing issues and expanding output formats. The user added capabilities to generate text with custom orientations, space widths, and various distorsions. Furthermore, they integrated support for generating text in multiple languages and added support for custom fonts and image backgrounds, including quasicrystal, picture and handwritten text.
nlpdatatext-recognitionrecognitiondataset
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.