Muhammad W.

I am an AI/ML Solutions Architect and data scientist with 10+ years of combined industry and academic experience across machine learning, LLM/RAG systems, predictive modeling, financial analytics, insurance analytics, NLP, computer vision, and advanced statistical analysis. My background includes research experience at CERN in Geneva, Switzerland, and Fermilab near Chicago, USA, where I applied machine learning to complex, high-volume scientific data. I have also co-authored 20+ research papers. In industry, I have worked across startups, enterprises, insurance, and GCC government/financial-sector environments. My recent work includes building LLM-based chatbot and decision-support systems on complex financial and insurance data, enabling business users to query structured data, analyze performance, understand risks, and generate decision-ready insights. I have worked with the Department of Finance – Abu Dhabi on system development and AI/ML use cases involving budget forecasting, expenditure tracking, anomaly detection, macroeconomic benchmarking, and chatbot-style decision support. In my current insurance role, I have built an internal AI chatbot over underwriting, pricing, claims, quotation, and financial performance data. Core expertise: ✔️ LLM/RAG chatbot development over structured and unstructured business data ✔️ AI agents and workflow automation for retrieval, analysis, summarization, and decision support ✔️ LangChain, LlamaIndex, embeddings, vector search, prompt design, and retrieval pipelines ✔️ Machine learning, predictive modeling, anomaly detection, and time series forecasting ✔️ Financial, insurance, government, and operational analytics ✔️ Model validation, data-quality checks, leakage detection, business-rule validation, and output testing ✔️ Dashboards and analytics apps using Python, Streamlit, Power BI, Tableau, and SQL ✔️ Python, PyTorch, TensorFlow, scikit-learn, pandas, PySpark, SQL, AWS, and Azure ✔️ Advanced statistics, regression, classification, clustering, XGBoost, Random Forest, LSTM, and neural networks What differentiates me is the combination of scientific rigor, GCC financial-sector experience, and hands-on AI/LLM delivery. I do not only build models or chatbots — I validate the data, challenge assumptions, test outputs, and ensure the final solution is reliable, explainable, and useful for real business decisions. Feel free to reach out to discuss project scope, availability, and rates.