Naoufal SAADI

AI & Data Consultant @ Effyis Group

About

Data Scientist & Machine Learning Engineer | Passionate about building scalable AI systems and data-driven solutions. Experience with RAG (90% recall), predictive analytics, and real-time pipelines (Kafka, Spark). Open to opportunities in Data Science · ML Engineering · Data Engineering. I thrive at the intersection of AI and engineering, blending hands-on problem solving with measurable impact: 🚀 Built a Retrieval-Augmented Generation (RAG) system on Hugging Face that achieved 90% recall, boosting precision in scientific knowledge retrieval. 📊 Developed and deployed predictive forecasting models, improving sales strategy and decision-making. ⚡ Optimized real-time data pipelines with Apache Kafka & Spark Structured Streaming, enabling near-instant analytics. 🧠 Fine-tuned deep learning models to better understand behavior and improve prediction accuracy. Curiosity drives me to keep learning and pushing boundaries in LLMs, real-time analytics, and applied machine learning research. 🔍 I’m eager to bring this mix of technical expertise and impact-driven mindset to a Data Science, ML Engineering, or Data Engineering role where I can contribute to building innovative, scalable solutions.

Country

Morocco

City

Rabat

Industry

Information Technology & Services

Skill

agentic ai, Business, E-Commerce, Finance, Business Intelligence (BI), Business Insights, Dashboards, Analytics, Chatbot Development, Fine Tuning, Communication, PyTorch, Learning, Neuroscience, Engineering, Neural Networks, Scikit-Learn, Programming, Artificial Intelligence (AI), Java

Experience

AI & Data Consultant

Effyis Group

2025-10 - Present · 1 yr

Casablanca-Settat, Morocco

Data Scientist

INTELCIA

2025-2 - 2025-8 · 7 mos

Casablanca, Casablanca-Settat, Morocco

Engineered a multi-agent LLM framework combining a domain-specific knowledge graph with RAG to turn natural-language questions into optimized BigQuery SQL, cutting complex-query latency from 2 min → 1 min (-50%) and routine queries from 45 s → 30 s (-33%). Automated the insight loop—text-to-SQL → result analysis → visualization using Python, LangChain agents, OpenAI function calling, and BigQuery UDFs, shrinking dashboard build time from hours to minutes. Delivered an interactive Chainlit interface with personalized auth and memory for granular answers, follow-ups, real-time visualizations, and streamlined review of generated queries and feedback.

AI Research Intern

3D Smart Factory

2024-7 - 2024-10 · 4 mos

Mohammedia, Casablanca-Settat, Morocco

- Developed a RAG system using Hugging Face datasets and an arXiv corpus of scientific papers, achieving a 90% recall and strong precision in answering complex scientific queries. - Enhanced the RAG pipeline into an advanced system by integrating user profiles (researcher, student, professor), delivering personalized, high-precision answers tailored to each user's background. -Led the deployment process, improving clarity and direction in responses while guiding users toward relevant reading materials.

Data Science Intern

Ecole Nationale Supérieure d'Informatique et d'Analyse des Systèmes - ENSIAS

2023-7 - 2023-9 · 3 mos

Rabat, Rabat-Salé-Kénitra, Morocco

Prepared and cleaned datasets for analysis, ensuring data quality and reliability. Explored machine learning models to understand behavior and improve predictive accuracy.

BI& Data science Intern

Technocolabs Softwares Inc.

2023-7 - 2023-8 · 2 mos

Indore district, India

Developed and deployed a sales prediction project, leveraging machine learning algorithms to forecast sales trends and optimize business strategies. Analyzed and manipulated data using Excel, implementing preprocessing techniques to enhance quality and reliability. Visualized data and created interactive dashboards using PowerBI, contributing to a 20% increase in data accessibility.