I’m Meriem, a Machine Learning researcher and Data Scientist currently pursuing a
Master 2 in Machine Learning & Data Science at Université Paris Cité, France.

My interests lie in Deep Learning, LLMs, IoT data pipelines, multivariate signal analysis, and applied AI in real-world environments.
I enjoy building systems that combine research depth with industry-grade engineering.

Here’s a snapshot of my path so far:

• 🎓 Master 2 – Machine Learning & Data Science, Université Paris Cité (2025 – current)
• 🎓 Master 2 – Artificial Intelligence, ENSIA (ESI Alger) (2021–2024)
• 🤖 Data Scientist – LLM-based Document Intelligence, TotalEnergies (Paris, 2025)
• 📡 R&D Data Scientist – IoT Medical Sensors, CERIST (2024)
• 📈 Data Scientist – Forecasting & Reporting, Ooredoo (2023)
• 📝 Published IEEE Research (2024) – ECG Arrhythmia Detection with DL + Gaussian Models
• 🔬 Builder of ML, DL & IoT end-to-end prototypes for real-world use cases

My research focuses on:

• Deep learning for biomedical signals
• Transformer-based and hybrid CNN/LSTM architectures
• Multivariate IoT sensor analysis & anomaly detection
• Gaussian modelling & probabilistic approaches in ML
• Efficient ML systems for real-time monitoring

Integrating Deep Learning for Comprehensive Detection and Optimization of ECG-Based Arrhythmia Using Gaussian Function Model
Authors: Sahar Boulkaboul, Meriem Baha, Medjda Rihab Slimani
📍 IEEE HONET 2024 — Doha, Qatar
🔗 https://ieeexplore.ieee.org/document/10822889

Keywords:
CNN • LSTM • Transformers • Channel Attention • Gaussian Functions • ECG Signals
This work proposes a hybrid deep learning pipeline that integrates Gaussian beat modelling with attention-based neural networks for robust arrhythmia detection.

• Automated analysis of PDF, PPTX and Excel documents at scale
• OCR pipelines (Tesseract / PaddleOCR)
• Retrieval-based enhancement using LLMs
• Built a Streamlit dashboard for KPI monitoring
  Stack: Python, LangChain, OCR, GCP, Dataiku, Streamlit

• Built ML pipelines for ECG/SpO2 sensor data
• Automated data quality checks on Raspberry Pi
• Deep learning for abnormal pattern detection
  Stack: PyTorch, FastAPI, MQTT, MongoDB, Streamlit

• LSTM prediction models for call volume forecasting
• SQL pipelines for reusable datasets
• Interactive reporting dashboards
  Stack: Python, SQL, TensorFlow, Streamlit

• Multimodal RAG assistant (vector DB + embeddings + LLM)
• Data engineering workflows for structured/unstructured data
• ML monitoring
• End-to-end deep learning training pipelines

• 📝 Published IEEE Author — HONET 2024
• 🎓 AWS Solutions Architect – Associate (2024)
• 🔬 Built multiple research-backed DL & IoT systems
• 📊 Designed complete ML pipelines deployed in real settings
• 💡 Strong ability to bridge engineering with applied research

• Deep learning for complex signal data
• ML pipeline engineering & automation
• Real-world LLM applications
• Cloud-based ML deployment

• Applied Machine Learning research
• Large Language Models (LLMs) & Generative AI systems
• RAG pipelines & enterprise search solutions
• End-to-end ML & Data Engineering workflows
• Scalable AI/ML systems for production
• Advanced Deep Learning architectures (CNN, LSTM, Transformers)

💼 LinkedIn: https://www.linkedin.com/in/Meriem-Baha
💻 GitHub: https://github.com/MERYX-bh
📧 Email: meriembaha2611@gmail.com

Thanks for visiting my profile! ✨
Always happy to discuss ML, AI systems, research and real-world applications 🚀