Close this
Close this

Sharmisha Yelapati

Development
New Jersey, United States

Skills

Data Science

About

SHARMISHA YELAPATI's skills align with IT R&D Professionals (Information and Communication Technology). SHARMISHA also has skills associated with Database Specialists (Information and Communication Technology). SHARMISHA YELAPATI appears to be a low-to-mid level candidate, with 4 years of experience.
View more

Work Experience

Data Scientist

STERIS
September 2023 - Present
  • Architected a cutting-edge NLP solution utilizing the NLTK library, revolutionizing data processing and pattern recognition accuracy to drive an improvement in sentiment analysis precision, enhancing customer insights and decision-making processes. Exploit Attention and Transformer models, such as BERT and GPT, achieving a 25% improvement in natural language processing tasks. Led cross-functional workshops with key stakeholders and domain experts to define BERT application parameters; resulted in a 20% reduction in customer churn rate by implementing personalized recommendation algorithms based on user behavior data. Manage OpenCV for image preprocessing, resulting in a 15% enhancement in data quality for computer vision tasks. Implemented Amazon EMR to streamline distributed data processing, resulting in a 50% reduction in processing times, 40% faster execution of complex analytics, and a 25% increase in overall system performance. Generate and execute a serverless architecture leveraging AWS Lambda and Amazon API Gateway, resulting in a 40% reduction in infrastructure costs and a 50% improvement in system scalability.

Data Scientist

Macro Software Solutions
August 2019 - July 2022
  • Led the development and implementation of Convolutional Neural Networks (CNNs) for image classification tasks, resulting in an average accuracy improvement across multiple datasets, optimizing model efficiency and accuracy. Orchestrated the application of advanced statistical modeling techniques with XGBoost in Python, driving a 25% improvement in customer churn prediction accuracy and enabling targeted retention strategies. Incorporated PyTorch to construct graphs, thereby enhancing model accommodating and reducing progress iterations by 26%. Accelerated data preparation by 30%, leading to a 5% boost in model accuracy, by designing customized text cleaning and preprocessing tools that effectively handled domain-specific language patterns. Engineered an automated customer response system by leveraging NLTK module in Python, resulting in a 20% reduction in response time and increased operational efficiency by 30%. Designed and developed interactive Power BI dashboards and reports for 50+ business users, facilitating data-driven decision making. Optimized DynamoDB tables for specific use cases, resulting in a 23% increase in query performance and a 19% reduction in storage costs. Accomplished the integration of AWS services with on-premise resources, creating a hybrid cloud environment that increased flexibility and compact operational costs by 29%.

Education

Coursera

Certificate

Rowan University

Master of Science in Data Science

CMR Institute of Technology

Bachelor in Mechanical Engineering