• Led end-to-end NLP projects, including data collection, model training, evaluation, and deployment, driving significant operational efficiency improvements.• Improved classification accuracy through Text Classification and Sentiment Analysis, delivering key business insights.• Increased process automation efficiency by 81% via advanced sentiment analysis and topic classification using Hugging Face models.• Fine-tuned BERT and MPNet models in AWS Sagemaker, achieving substantial improvements in classification accuracy.• Optimized prediction speed by developing a similarity-based model using FAISS for fast indexing of embeddings.• Enhanced model performance by integrating rule-based, machine learning, and similarity-based models, leading to notable accuracy gains.

• Developed a prototype of customized deep learning models using CNN architecture for predicting COVID-19 from Chest X-rays.• Employed data augmentation and transfer learning techniques to enhance model performance, overcoming challenges due to the lack of open-source datasets.• Leveraged transfer learning to build a robust COVID classifier, utilizing the strengths of VGG-Net and ResNet models to improve prediction accuracy.

• Conducted extensive research on methods for detecting brain tumors from MRI images, laying the groundwork for model development.• Compiled and prepared datasets by gathering relevant data and applying image processing techniques using OpenCV and other Python libraries.- Implemented a variety of machine learning techniques, including CNNs, Random Forest, and XGBoost, fine-tuning model parameters for optimal performance.• Enhanced CNN model performance through meticulous fine-tuning, evaluated models on multiple metrics, visualized results, and thoroughly documented findings, contributing to advancements in medical imaging.