Naghul Meera Shaik

Data Scientist @Credit Suisse

New York, NY, US

Spearheaded a sentiment analysis pipeline utilizing LangChain to classify customer feedback, resulting in a 10% increase in customer satisfaction and informed business strategy.• Leveraged Tableau to perform customer segmentation based on key attributes, driving a 10% improvement in the effectiveness of targeted marketing campaigns.• Leveraged transfer learning to fine-tune a pre-trained BERT model for sentiment analysis on a smaller, task-specific dataset, cutting training time by 50% and boosting the F1 score by 10%.• Evaluated NLP model performance using BLEU, ROUGE, and perplexity metrics to gauge translation quality, assess summarization effectiveness, and measure language generation fluency.• Engineered predictive models utilizing Python libraries (NumPy, Scipy, pandas, Scikit-learn) and Spark, resulting in an average accuracy enhancement of 18% across all models, revolutionizing data-driven decision-making processes.• Optimized TensorFlow\'s high-performance computing capabilities to design and train complex deep learning models, achieving a 30% reduction in training time compared to traditional frameworks.• Applied advanced transfer learning techniques with pre-trained PyTorch models, accelerating training processes by 50% and enhancing model performance on target tasks.

Pace University - Seidenberg School of Computer Science and Information Systems

Master of Science - MS