Artificial Intelligence Engineer
Rawan Sadek is an Artificial Intelligence Engineer with expertise in developing RAG chatbots, optimizing web scraping operations, and enhancing computer vision systems. She focuses on creating reliable and scalable AI solutions, leveraging deep learning and natural language processing to solve complex problems.
Expertise in developing and deploying AI solutions, including RAG chatbots, predictive modeling, and deep learning applications.
Skilled in text cleaning, feature extraction (TF IDF, stemming, lemmatization), and building models for tasks like job fraud detection and natural language interaction.
Proficient in applying computer vision techniques such as contrastive learning, pattern extraction, and data augmentation to enhance detection systems and image classification.
Experienced in leading large-scale web scraping operations using Python/Scrapy, ensuring high data accuracy and optimizing crawl efficiency.
Extracted shapes from heavily pixelized, noisy images using contour detection, noise reduction, and k-means clustering. Calculated areas, curvature, and angles while addressing distortions and removing excess corners for more accurate measurements.
Impact: More accurate measurements from challenging image data.
Built a contrastive learning model for distracted driving detection, achieving high accuracy across varied driving scenarios. Applied targeted augmentation to improve robustness under low-light, varied postures, and changing environments.
Impact: Reduced model size by 95% (25M → 1.25M parameters) with SqueezeNet, enabling real-time inference.
Developed a job fraud detection model, achieving 98.3% accuracy with TF IDF and logistic regression. Improved feature extraction using stemming and lemmatization techniques.
Impact: Achieved 98.3% accuracy.
Trained Xception, CNN, and ResNet models on 1,700 sports images with augmentation.
Impact: Achieved 93.1% accuracy on 650 sports test images.
Researching and developing a Retrieval-Augmented Generation (RAG) chatbot at Reva Group to enable accurate natural-language interaction with rental data.
Led the Area of Responsibility (AOR) for large-scale web scraping with Python/Scrapy at Reva Group, ensuring high data accuracy and reliability.
Designed and developed a RAG-based chatbot as the sole developer to streamline access to internal documentation during an internship at Valeo.
Graduated with First-Class Honours in B.Sc. in Computer Science from the University of East London.
Completed "Neural Networks and Deep Learning" certification.
Enhanced Occupant Monitoring System by analyzing video samples to identify and resolve detection issues, reducing false positives and improving reliability at Valeo.