Rana Mohamed

Designed an assistive robotic-hand control system using OpenCV and MediaPipe with Raspberry Pi, achieving 92% grasp success in 100 lab tests and sub-100ms latency for gesture and eye-tracking control.

Specialized in object detection and gesture recognition using OpenCV and CNNs; contributed to real-time image processing projects with high accuracy in dynamic environments.

Completed comprehensive ML engineering training covering algorithms, preprocessing, model optimization, MLOps and deployment; developed and deployed Azure ML models and delivered a capstone project focused on scalable AI solutions.

Built transformer-based NLP pipelines for speech-to-text, summarization, and topic classification using TensorFlow, Sumy and NLTK, delivering coherent multi-task text generation with high accuracy.

Studied foundational big data analytics and distributed storage concepts; worked with Hadoop, MapReduce, and HDFS on telecom-focused hands-on projects.

Built a real-time face recognition system achieving 95% accuracy on a custom dataset using OpenCV and DeepFace, optimizing pipelines for diverse lighting and angles for security and attendance use cases.

Pursuing deep learning coursework and projects in TensorFlow and PyTorch focused on NLP and computer vision pipelines, data collection, and model training.

Completed training on Python fundamentals and scientific libraries including NumPy, Pandas, and Matplotlib, applying them to data processing tasks and small projects.

Developed a CNN-based Sign Language recognition model using TensorFlow and Sign Language MNIST to enable real-time ASL letter recognition through preprocessing and model optimization.

Grade: 3.55

Activities and societies: Projects in face recognition, sign language recognition, assistive robotics (Hand Vision), and NLP text-generation and summarization.

Pursuing a Bachelor of Science in Artificial Intelligence with coursework and projects in NLP, computer vision, and machine learning; current GPA 3.55 and expected graduation in 2026.