About

Mohamed Sharafath is a highly skilled Senior Research Engineer with over 4 years of experience building AI-driven software systems. He has extensive experience in the healthcare revenue cycle automation domain and is proficient in Machine Learning, Deep Learning, NLP, Scala, and Python. In his current role as a Senior Research Engineer II at buddi.ai, Mohamed has designed and built technical solutions for AI/ML use cases including data engineering, EDA, feature engineering, algorithm selection, model tuning, model evaluation, insights, and recommendations. He also has extensive experience in AI/ML model deployment and monitoring. Mohamed has developed a multi-level Deep learning-based automated medical coding solution to classify documents among 70k classes. With this solution, manual coders spent 22% less time coding a clinical text. He has also designed and developed an engine with custom NER, Relationship Extraction, and Coreference Resolution models for extracting medical entities and their relationships in clinical texts. He combined deep learning solutions with classical NLP techniques for high accuracy. In addition, Mohamed has built, tuned & deployed tree-based(Xgboost) models to predict defective semiconductor components ahead of time for Industry 4.0 with a 90%+ ROC-AUC score. He orchestrated MLOps pipeline using KubeFlow for continuous model training, evaluation, and monitoring. He has also performed extensive Data Manipulation & Feature Engineering to produce consumable datapoints. Mohamed holds a Bachelor of Engineering (BE) in Electrical and Electronics Engineering from Anna University. He is a Research Scientist with expertise in Deep Learning, ML, NLP, AI, Python, Keras, BERT, Data Engineering, Pytorch, Scala, Information Retrieval, XGBoost, CI/CD, SQL, AWS, Sentiment Analysis, Corpus, and Data Scientist. Overall, Mohamed has relevant experience of 6.62 years and is enthusiastic about building AI-driven software systems that solve real-world problems.