Ayoub D.

I’m Ayoub, an AI/ML engineer specialized in building and deploying advanced language and vision systems. Core stack & focus areas: LLMs & VLMs: Custom LLM development, fine-tuning with LoRA/PEFT, multimodal architectures, transformer-based models (ViT, DETR, SAM, Florence) Retrieval-Augmented Generation (RAG): End-to-end search & retrieval pipelines with LangChain, vector databases (FAISS), and domain-adapted embeddings for intelligent document understanding Vision & NLP Systems: OCR pipelines (VLM ,Tesseract, EasyOCR), document parsing, entity extraction, structured data generation, and computer vision models (YOLO, SAM, OMNIPARSER) End-to-End AI Pipelines: Automated systems for data scraping, preprocessing, CV/NLP inference, metadata extraction, and API-based orchestration Backend & Deployment: Python (PyTorch, FastAPI), REST API development, Streamlit apps, and CI/CD workflows I specialize in production-ready AI solutions—from research and prototyping to fully deployed pipelines—particularly in LLMs, multimodal models, and intelligent document-processing systems. Open to projects involving RAG systems, LLM/VLM fine-tuning, scalable AI automation, and multimodal intelligence.