محمد طیب

I’m an AI Engineer focused on turning complex requirements into practical, production-ready GenAI systems. My work centers on RAG, LLMs, NLP, and computer vision, with a strong preference for solutions that feel fast, reliable, and genuinely helpful for real users.

In my current role, I led end-to-end development of a multi-module AI platform for industrial safety management. I integrated Retrieval-Augmented Generation, tool-calling agents, and FastAPI-based microservices to deliver risk assessment, multi-prompt assistance, “Ask” functionality, and legal compliance workflows.

I also built a modular AI system with multiple RAG pipelines and agents that power legal query chatbots, document/project QA, multi-document summarization, construction drawing analysis, and Excel automation. I designed automated pipelines for extracting and transforming content across PDF/DOCX/XLSX and images, including metadata parsing and FAQ generation with inline citations—while adding streaming, caching, concurrency, and persistent chat history.

Beyond production systems, I develop specialized capabilities like document analysis with real-time retrieval and rerankers, and zero-shot voice cloning using MaskGCT with transcription via Whisper. I’ve also shipped focused applications including a safety gear detector with YOLOv8 + Streamlit, and an AI interviewer that generates interview questions and records and transcribes responses using Gemini and Deepgram. I value clean engineering, measurable quality, and AI experiences that stay grounded in evidence and user context.