Sakshi Srivastava Email and Phone Number

- Manipulated a million transaction records to create 360-degree customer profiling for sales predictions for ‘General Motors’ and used Geographic Information Systems and data analytics on real time IOT data for boosting the dealership performance by 30%- Built a predictive analytics platform that generates insights into several core business strategies, including anticipating market demand for autonomous vehicles and forecasting the annual sales.

• Created a 5G widget 'Network Decision Platform' for the 'Centre for Data & Insights' initiative in 'Comms, Media and Telecom' on Tableau by utilizing third-party data to enhance network operator’s data resulting in a significant enhancement of network performance and increased customer satisfaction scores from 65% to 90% within a year.• Implemented a data-driven strategy for ‘British Telecom’ to reduce customer churn by 20% caused by the discontinuation of 3G services, including customer segmentation & analysis using Python to enable upselling and cross-selling opportunities and data model creation for personalized recommendation.• Collaborated with product managers, engineers, data analysts and stakeholders to identify key business problems and develop data-driven solutions and mentored 3 team members under me.• Leveraged supervised and unsupervised learning algorithms like linear regression, logistic regression, random forests, and k-means to build predictive models resulting in a 15% increase in customer engagement.• Orchestrated ETL and automated CI/CD data pipelines deploying models in production on GCP cloud platforms using Terraform, Airflow and Big data Query to streamline data processing and reduce manual effort by 30%.• Designed a scalable data architecture that aggregated data from multiple sources and reduced data processing time by 50% processing huge data to uncover key insights, resulting in a 15% increase in revenue.

- Enhanced a self-service business analytics application IMAnalytics that enables operations to seamlessly interact with the product IMActions data and provide instructions like investment decisions & dividend declaration to custodians.• Generated weekly and monthly analytical reports and insights for JPM, CITI, Nordea, and UBS, which were used to inform key investment decisions and improve overall performance leading to a 20% increase in client satisfaction and a 10% increase in assets under management.• Implemented a fraud detection model using SMOTE technique on a sample dataset, which resulted in a 75% accuracy rate and reduced false positives by 50%. This model was deployed in production and saved the company $100,000 per year in fraudulent transactions.