Lahoucine MKHARBECH

From raw solar streams to structured intelligence → Designed a Python–PostgreSQL pipeline seamlessly ingesting client API data for real-time anomaly detection. From detection to action → Built machine learning models that not only identify anomalies but also trigger automated alerts reaching clients across Europe, America, Asia, and Africa. From queries to conversations → Developed an AI-powered SQL agent, enabling teams to interact with the database using natural language, making insights instantly accessible. Each layer contributes to the same vision: transforming data into a living system of intelligence, automation, and accessibility.

At Rawa Services, I joined a team tackling a critical challenge: making sense of complex energy consumption data for clients across multiple regions. The journey started with messy, unstructured API data. I designed and deployed a Python-based pipeline integrated with PostgreSQL, ensuring seamless and reliable data ingestion. But clean data was only the beginning. The real mission was to create business value. I built an automation model that detected consumption anomalies (P10 data), reducing processing time by 80% and empowering clients to act faster. To make this innovation truly accessible, I developed a user-focused web application using Flask, Docker, and AWS — transforming technical outputs into practical tools. This experience taught me how to bridge the gap between data science and business needs, turning invisible patterns into insights that drive smarter decisions.

Automatisation d'analyse de la surconsommation d'énergie : python, data extraction, preprocessing, machine learning, clustering, interface utilisateur...