Ismail Baztami

– CAF Ticketing Support (WhatsApp): Developed an AI chatbot leveraging LangGraph for complex decision orchestration and Retrieval-Augmented Generation (RAG); automated over 300 tickets daily and significantly reduced average resolution time. – GenAI Candidate Selection: Collaborated with the HR department to develop an automated recruitment tool using LangChain and OCR to extract key information from CVs; implemented an LLM-based scoring system to rank candidates against job requirements, streamlining the initial screening process. – Analysis & Specifications: Conducted requirement gathering with end-users to define prospecting scenarios and authored detailed functional specifications for CRM integration. – Multi-Agent System Design: Architected a sophisticated voice-based multi-agent system using LangGraph, integrating Speech-to-Text (STT) and Text-to-Speech (TTS) modules with direct connectivity to internal telephony systems and the CRM via API. – ETL Pipeline & Orchestration: Engineered automated ETL workflows to extract, transform, and load system logs into Power BI, enabling real-time performance monitoring and data-driven insights.

Retail Inventory Forecasting: Analysis, visualization, and interpretation of sales data to optimize stock levels and reduce out-of-stock incidents, utilizing an automated data preparation pipeline and an XGBoost demand forecasting model.

– Modeling & MLOps: Engineered data preprocessing pipelines and evaluated multiple classification algorithms to identify benign vs. malignant cells; conducted extensive hyperparameter tuning and utilized MLflow for experiment tracking, ensuring the selection of the highest-performing model through rigorous performance metrics.