Vinithra Venkatesh

• Design & lead analytic solutions. Involves creating a comprehensive roadmap, hiring, training, and managing a team of Analysts.
• POC between the global business team and technical team to create robust, actionable data science and BI solutions while also working hands on to develop state of the art data science solutions.

Key Projects:

Led Migration of ELT pipelines from Nifi to Databricks:

• Contributed to cost reduction by 95% and improved error identification and resolution efficiency, achieving a 7x reduction in execution time.
• Orchestrated the conception, development, testing, and integration of multiple modules within stipulated timelines, ensuring a seamless migration to Databricks.
• Provided comprehensive training to the team on utilizing Databricks and PySpark effectively.

Spearheaded CRM Customer Profiling:

• Initiated a project to gain customer cohort-level insights, supporting targeted marketing strategies.

AI Innovation:

• Sentiment Analysis- Analyzed sentiment of customer reviews using Natural Language Processing and comparing results using generative AI.
• Virtual assistant- Developed a virtual assistant to help professionals handling various work-related tasks.
• Tested Hugging face sentence transformers and other Large Language Models and finally implemented using custom GPT 4 API

• Created and lead a specialized team to support CRM campaign managers.

Key Projects:

• Forecasting simulator - Built a forecasting product end to end to show how the future compensation values changes when manipulating drivers using time series forecasting models, regression and Power BI for the user interface.
• Campaign Evaluation framework - Built a comprehensive robust framework that will measure a campaign wholistically and provide a singular assessment criterion for a campaign. This enables campaign managers and sales heads with all necessary information to prioritize and take corrective action.

Key projects

• Involved in building and deploying Machine learning models for proactive customer contact across the bank’s product portfolio.
• Customer satisfaction driver model - Built driver model developed an engine to categorize drivers to prioritize them on impact.
• Second home proactive customer contact - Built Machine learning model to prioritize leads. Enabled DKK 56 million quarterly in business volume.
• Customer product journey - Using clustering and RFM techniques to understand the customers’ current product journey and predict their next best product.