Devi K Email & Phone Number

New Jersey, United States

Collaborated with Cinergy Technology, a growing financial services company, to optimize its financial transaction processing system. Employed cleaning techniques to ensure the data was clean and utilized data visualization libraries from Python such as Matplotlib, Seaborn, and ggplot. Utilized time series and ratio analysis for data analysis, as well as Power BI and Tableau for deeper analysis and accurate visualizations. Employed supervised machine learning techniques such as random forest and support vector machine for the fraud detection analysis and unsupervised ones such as K means clustering for anomaly detection such as unusual spending patterns. As well as Linear regression to identify the Cost optimization trends.Developed various client applications leveraging REST and SOAP APIs to gather, create, and update data across multiple agency management systems.Worked with platforms such as Salesforce to enhance data integration and streamline data processes for improved functionality and efficiency.Enhanced client and data management by designing solutions that facilitated seamless data exchange and integration.

Denton, Texas, United States

As a Graduate Student Assistant in Data Science, I was responsible for supporting various data-driven research projects and enhancing the university's data infrastructure. My key responsibilities included:Assisted in gathering large datasets from multiple sources, including databases, APIs, and web scraping techniques. Cleaned and transformed raw data to ensure its suitability for analysis using Python libraries such as Pandas and NumPy.Conducted in-depth statistical analyses to uncover trends and patterns in complex datasets. Developed predictive models using machine learning algorithms, such as Linear Regression, Logistic Regression, K-means clustering, and Decision Trees, with tools like Scikit-Learn and TensorFlow.Created comprehensive visualizations to communicate findings to non-technical stakeholders. Leveraged tools like Tableau, Power BI, Matplotlib, and Seaborn to design interactive dashboards and reports, enabling data-driven decision-making.Worked with SQL and NoSQL databases (e.g., MySQL, MongoDB) to manage, query, and optimize data storage. Assisted in designing and implementing efficient data pipelines for ETL processes, ensuring smooth data flow between systems.Collaborated closely with faculty and fellow researchers to define project goals, identify relevant datasets, and develop actionable insights. Provided technical support in interpreting data analysis results, contributing to academic publications and presentations.Developed scripts to automate repetitive data processing tasks, improving efficiency and reducing manual errors. Implemented continuous integration and deployment (CI/CD) practices using Git for version control, ensuring seamless project updates and collaboration.Maintained detailed documentation of data processing workflows, analysis methodologies, and codebases. Created user manuals and technical guides to facilitate knowledge transfer and support ongoing research initiatives.

India

At Golok Soft Global, I worked as a Data Scientist for Berymo, an on-demand ride-sharing client, where I transformed data into actionable insights, enhancing business operations and customer experience.Developed predictive models for driver availability and demand forecasting using Python, leading to a 20% reduction in wait times and an overall increase in ride completion rates.Built and optimized ETL data pipelines using SQL and Python, automating data cleaning and processing for real-time analytics, which improved data accessibility for Berymo’s operations team.Implemented customer segmentation using clustering techniques (K-means, hierarchical clustering), which informed targeted marketing strategies, increasing user engagement by 15%.Designed dashboards in Power BI to visualize key metrics like trip frequency, churn rate, and average wait time, allowing stakeholders to make data-driven decisions.Collaborated closely with Berymo’s product and engineering teams to align data solutions with business goals, ensuring measurable impact across user retention and revenue metrics.