I specialize in machine learning development, with hands-on experience in building and deploying ML models. Proficient in Python, C++, C#, HTML5, and CSS, I also have a strong foundation in project management.
linkedinWith a solid technical background and certifications in PyTorch, networking, and data manipulation, I enjoy tackling complex challenges. My portfolio includes projects like LipLingo, sales forecasting and sentiment analysis, reflecting my approach to using technology for practical, data-driven insights.
LipLingo is a deep learning model that enhances automatic lip reading for speech recognition and aiding the hearing impaired. Built on a modified LipNet architecture, it uses CNNs for spatial features, bidirectional LSTMs for temporal modeling, and dense layers for prediction. With a robust TensorFlow pipeline and improved accuracy over existing models, LipLingo advances accessible communication technology.
View on GitHub
Performed professional sentiment analysis using Python, pandas, Matplotlib, NLTK, and scikit-learn. Extracted nuanced emotions from text for insightful decision-making.
View on GitHub
Conducted a sales forecasting project using Linear Regression in Python, employing NumPy, pandas, and Matplotlib. Evaluated model performance with metrics like mean squared error, mean absolute error, and R2 score, providing valuable insights for informed decision-making in sales analytics.
View on GitHubA comparative study on data augmentation and transfer learning techniques for sentiment classification on small datasets. Published in the IEEE Transactions on Artificial Intelligence.
Read PublicationImplementation of a federated learning system using the Flower framework with a custom server strategy. Targeted at secure AI for connected industries.
Read Publication