Tasnime Lahmar

The project focused on developing a Multi-Agent Anomaly Detection System for System Logs, combining Large Language Models with deep learning to enhance anomaly detection across diverse infrastructures. At its core, the project involved: 🔹 Designing a modular multi-agent framework using LangGraph, with each agent dedicated to a specific function: orchestration, detection, and visualization. 🔹 Integrating LLMs to coordinate detection tasks and assist with root-cause analysis. 🔹 Applying meta-learning and unsupervised techniques to improve generalization across systems. 🔹 Leveraging Generative AI to automatically generate a dynamic Power BI dashboard for real-time visualization of anomalies and KPIs. 📈 This project bridges the gap between observability and intelligent automation, aiming to support real-time decision-making and scalable log analysis.