Sakshi Gupte Email & Phone Number

Dartmouth, MA, US

• Facilitated Agile ceremonies, including sprint planning, daily stand-ups, sprint reviews, and retrospectives, to ensure consistent project delivery and foster a culture of continuous improvement in software development.
• Designed and implemented a predictive analytics tool that segmented customers based on risk categories, leading to a 15% reduction in claims and enabling the creation of more effective, data-driven marketing strategies.
• Optimized marketing strategies by tailoring campaigns to customer risk profiles, reducing expenditures by 20% and increasing conversion rates by 25%, ultimately enhancing customer engagement and ROI.
• Developed and deployed a web-based application utilizing Docker for containerization, AWS ECR for image storage, and EC2 for hosting. Streamlined deployment using CI/CD pipelines and managed source control with Git.
• Integrated advanced Retrieval-Augmented Generation (RAG) systems using LLMs, Langchain, and Vector Databases, allowing stakeholders to access simplified, yet detailed, reports for improved decision-making across.
• Applied statistical validation techniques and SHAP analyses to insurance risk assessments, reducing prediction errors by 4% and increasing the accuracy and relevance of insights for underwriting decisions.

Mumbai, Maharashtra, India

• Performed comprehensive testing and validation of data science models in accordance with Waterfall methodology, ensuring the models met stringent accuracy and reliability standards prior to deployment.
• Conducted A/B testing to assess the performance of marketing campaigns and product features, leading to actionable insights that improved overall campaign effectiveness and customer engagement.
• Optimized large-scale data processing workflows using Apache Spark, reducing processing times by 40% and enhancing the overall throughput and scalability of data handling operations.
• Utilized machine learning techniques such as Naïve Bayes, Logistic Regression, Random Forest, and XGBoost to predict loan default probabilities, resulting in improved risk management strategies with increased.
• Enhanced model accuracy by 20% through the application of deep learning frameworks, including TensorFlow and PyTorch, effectively improving the performance of predictive models used for credit risk analysis.
• Automated complex data pipelines using Azure Data Factory and Azure Databricks, significantly increasing the efficiency of data engineering processes and minimizing manual intervention and potential errors.

Master Of Science - Ms, Data Science, 3.6

Hsc, Science

Activities and Societies: Dance Group

Ssc