Freelancer profile translated to English.
Back to original languageAbout Laziz
I help companies design and deploy artificial intelligence solutions to automate document analysis, information extraction, and complex data structuring.
Specializing in artificial intelligence through a CIFRE thesis in Document AI, Computer Vision, OCR, Deep Learning, and Generative AI, I have worked on industrial challenges related to automatic document processing at Malakoff Humanis. My expertise covers model design, training, evaluation, error analysis, and the implementation of prototypes usable by business teams.
I can work on projects involving document classification, OCR, information extraction, table recognition, form analysis, unstructured data structuring, computer vision, multimodal models, LLMs, RAG, and document process automation.
My deliverables can include proof of concepts, Python pipelines, trained or adapted models, analysis notebooks, APIs, evaluation reports, benchmarks, technical documentation, and recommendations for industrializing an AI solution.
French
Native or bilingual
English
Fluent
Can work on-site
Paris (up to 50km)
Experience
- LITIS & Malakoff HumanisData Scientist — CIFRE Thesis in Document AIFebruary 2023 - May 2026 (3 years and 3 months)Paris, FranceCarried out an industrial CIFRE thesis in artificial intelligence applied to automatic document processing, combining Computer Vision, NLP, OCR, generative models, model evaluation, and industrialization-oriented constraints.Designed end-to-end models for document analysis, particularly for OCR, information extraction, and complex table understanding. Worked on multimodal approaches capable of jointly predicting the text, structure, and layout of documents.Contributed to several industrial Document AI projects: document classification, named entity extraction, information extraction from dental estimates, invoices, social security reimbursement statements, and bank account details (RIB).Developed data and evaluation pipelines for synthetic document generation, annotation automation, model benchmarking, error analysis, visual inspection, and preparation of internal AI demonstrators.Evaluated OCR, VLM, LLM solutions, and hybrid pipelines for document understanding tasks, including routing strategies, deterministic validation, and programmatic verification for arithmetic or consistency checks.Supervised interns and apprentices on Document AI projects, providing support for technical choices related to GPU/cloud environments, particularly AWS EC2 and initial SageMaker experiments.
- Malakoff HumanisData ScientistMarch 2022 - July 2022 (4 months)Paris, FranceDeveloped computer vision and NLP models for fraud detection, inconsistency analysis, and anomaly identification in document images.Prepared machine learning models for industrial use through data preprocessing, validation, optimization, and production-oriented evaluation.
Recommendations
Be the first to recommend Laziz
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- PhD in Computer ScienceUniversity of Rouen Normandy2026Doctorat en informatique
- Master in Data ScienceUniversity of Rouen Normandy2022Master en Data Science