Freelancer profile translated to English.
Back to original languageAbout Salim
Expert in documentary AI with 5 years of experience — from fraud detection to document automation, I design diverse (Vision, RAG, Agent, Chatbot) and scalable AI systems that go into production on both cloud providers and on-premise. I transform complex documents into actionable data.
French
Native or bilingual
Can work on-site
Paris (up to 50km)
Experience
- Fleet360AI & ML Engineer / CDOJanuary 2025 - Today (1 year and 7 months)Fleet360 is an entrepreneurial project. I work as CDO and AI & ML Engineer. We design a SaaS solution to automate business processes and improve team productivity in car garages, dealerships, and parking lots.– Leading the design of a cloud application (GCP) for fleet document management.– Defining the data strategy and overseeing the project's technical architecture.– Developing the application on Streamlit and integrating advanced unstructured document processing.– Implementing OCR solutions and multimodal Vision models for document extraction and parsing.Stack: Python, Streamlit, OCR, Multimodal Vision Models, Document Parsing, GCP (Cloud Run, Cloud SQL, Vertex AI, Cloud Storage, Pub/Sub), Postgres
- EY, Numérobis,AI engineerJanuary 2026 - Today (7 months)As part of an investigation, a team of lawyers is conducting an in-depth review of documents. To effectively identify high-risk documents, I design and implement a solution based on a RAG approach. The infrastructure is entirely on-premise to ensure data confidentiality.– Processing unstructured documents and leveraging internal documentation.– Setting up a PostgreSQL database for email storage.– Implementing an agent to query the PostgreSQL database.– RAG architecture with indexing and semantic search on a Qdrant base.– Identifying document type and content for sorting processable data.Stack: Python, Streamlit, Multimodal Models, RAG (indexer, retriever), GitHub, VS Code, LangGraph, PostgreSQL, Qdrant
- EY, Produit FirtreeAI engineerOctober 2025 - January 2026 (3 months)Design and implementation of a RAG tool for corporate audits subject to the Sapin II law. The tool allows risk mapping, identification of non-compliance areas, and strengthening risk prevention for audit and compliance teams.– Processing unstructured documents and leveraging internal documentation.– Benchmarking and implementation of OCR solutions.– RAG architecture with indexing and semantic search.– Implementing LLMs to query the RAG and optimizing queries.– Testing RAG robustness on a dataset.– Automating the drafting of a standardized PowerPoint report for audit teams.– Implementing a fact-checker to highlight inconsistencies.Stack: Python, Streamlit, OCR, Multimodal Models, RAG, Azure (AI Foundry, AI Search, Blob Storage), GitHub, VS Code, Python-pptx
Recommendations
Be the first to recommend Salim
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
- Engineering DegreeÉcole Supérieure d'Ingénieur Léonard de Vinci (ESILV)2020Diplôme d'Ingénieur