A M Email & Phone Number

● Developed Spark Jobs to enrich data from multiple data stores like MsSQL, Cassandra, Oracle.● Cleaned and Preprocessed data to ensure its accuracy, consistency and ready for analysis.● Performed exploratory analysis to gain insights into the data, identifying data and uncover Potential problems.● Applied domain knowledge to select and transform features effectively. ● Developed and trained machine learning models using libraries like scikit-learn, TensorFlow, or PyTorch. ● Evaluated model Performance using appropriate metrics and techniques such as cross-validation.● Optimized model parameters to improve performance through techniques like grid search or random search.● Collaborated and communicated with teams and stake holders using JIRA.● Developed Predictive models to forecast future outcomes and trends.● Used Power BI for data visualization and presented analysis reports to the concerned departments.● Used GitHub for version control and code sharing with team members.

• Responsible for collecting and pre-processing large and complex Medicare/ Managed Care datasets to prepare them for analysis using Jupyter Notebook.• Used Python and R to perform data cleansing, transformation, and filtering such as identifying outliers, missing values, and invalid values.• Retrieved data from S3 and MySQL Server, Wrote PySpark SQL to retrieve, query and process structured data using Spark.• Developed Spark ETL jobs for moving data across multiple data stores/Filesystems and ran the jobs onAWS EMR.• Analyzed healthcare data, including claims data, clinical data, Provider, Prior Auth and demographic data by Box Plot, Scatter Plot and Histograms.• Identified patterns and trends in healthcare utilization and costs, as well as to identify opportunities for cost savings and quality improvement by clustering, decision trees, and neural networks.• Collaborating with other State departments including finance, marketing, and IT, in order to develop and implement healthcare programs and initiatives such as AI Dashboard.• Conducting cost-benefit analyses of various state accounts interventions and programs, in order to identify those that provide the most value for the organization.• Used JIRA for setting up and configuring Jira instances, managing user roles and permissions, creating and managing projects and customizing workflows.• Communicating findings and recommendations to different State Accounts, communicate your including senior management, clinical staff, and external partners.• Responsible for ensuring the quality of healthcare data, including identifying and resolving data quality issues, and ensuring that data is accurate, complete, and consistent.• Used Pandas, NumPy, SciPy, and Scikit-learn in Python for scientific computing and data analysis.

Texas, United States

Collaborated with other departments in gathering project requirements. ● Retrieved data from S3, Snow Flake, Cassandra and MySQL Server, Redshift.● Wrote PySpark SQL to retrieve, query and process structured data using Spark. ● Developed Spark ETL jobs for moving data across multiple data stores/Filesystems and ran the jobs onAWS EMR.● Developed Spark jobs using Scala for faster real-time analytics and used Spark SQL querying.● Used Python to perform data cleansing, transformation, and filtering such as identifying outliers, missing values and invalid values. Worked on feature engineering and data visualizations by Scatter Plot, Box Plot and Histogram Plot for performing EDA using packages Matplotlib and Seaborn in Python. ● Implemented principal component analysis (PCA) to emphasize variation and bring out strong patterns in datasets, also make the datasets easier to explore and visualize. ● Assisted in developing and testing AI models for customer behavior prediction.● Collaborated and communicated with teams and stake holders using Jira.● Dealt with massive imbalanced datasets by different methods such as oversample minority class, under sample majority class, generate synthetic samples and tune the performance metrics. ● Performed feature engineering such as feature selection using Recursive Feature Elimination, feature normalization and label encoding with Scikit-learn preprocessing library. ● Improved the model accuracy by 3% and solved the overfitting problem by using Random Forest and Gradient Boosting. ● Applied Cross Validation for hyperparameter tuning on different classification models and comparing the performance among different models. Validated the machine learning classifiers using ROC Curves. ● Used Pandas, NumPy, SciPy and Scikit-learn in Python for scientific computing and data● Used GitHub for version control and code sharing with team members.

● Participated in meetings with different teams to understand the project needs and requirements.● Collaborate with product managers, engineers, finance, and operation team to understand the current situation and help key decision making. ● Developed Python scripts to facilitate data collection from MySQL server and Copied data into S3 buckets using Spark jobs running on EMR. ● Extracted and organized information from manually conducted cases and exported to structured data using Python with re (regular expression). ● Specified data types to reduce memory requirements. ● Analyzed the data by Exploratory Data Analysis (EDA), worked on missing values. ● Worked on feature extraction and in creating new features using Pandas package. ● Used Principal Component Analysis (PCA) in feature engineering to analyze high dimensional data.● Worked on outlier identification using Box Plot with Matplotlib, NumPy, and Pandas.● Performed customer segmentation models using K-Means and Gaussian Mixture Model clustering algorithms. ● Created Hive tables to load data as Parquet and ORC files for processing. ● Copied the ORC files to amazon S3 buckets using Spark for further processing in the Amazon EMR cluster. ● Developed Decision Tree model to identify key predictors for the models. ● Used a Random Forest classifier to check the booking volume changes with the booking price ranges to get the optimum booking price. ● Worked with data visualization tools in python like Matplotlib and Seaborn.● By using JIRA, we improved project efficiency, enhanced team collaboration, increased productivity and successfully delivered the project.● Used Tableau for data visualization and presented the analysis reports to the concerned departments.

Hyderabad, Telangana, India

Worked on predicting the student’s chance of admission into state government schools based on their merit and coding skills. ● Analyzed the students by their performance and gave recommendations on the areas of improvement.● Longitudinal dataset was built using coding skills, scores, and student information.● Performed validation checks to find whether all the columns are present, and all are of similar data types. ● Implemented clustering for better understanding of the data and we divided into different groups by DBSCAN.● Implemented various supervised classification machine learning algorithms like logistic regression, random forests, SVM, KNN, Feed-forward ANN and handled class imbalance in the data using SMOTE. ● We found the accuracy by Confusion matrix. ● To increase the performance, we implement hyper parameter tuning.

Hyderabad, Telangana, India

● Responsible for the development and handling of the OKTA application, an enterprise-grade, identity management service compatible with many applications including cloud. ● Assisted users with the application registration process, communicate, share files and media. ● Used Okta's UI to add or remove users, modify profile and authorization attributes, and to quickly troubleshoot user sign-in issues. Okta gives you one place to manage your users and user data. Users can be synced from a variety of services, third party apps, and user stories.● The Admin Console contains predefined reports, system log filters, and notification tools to achieve most of these tasks. If you have external commercial or custom monitoring tools, you can integrate them with your Okta org. Okta sends the integrated tools a continuous flow of event logs or alerts for specific configured events.