VIJAYASHRI RANGASWAMY

M: 858-***-**** / adjlyb@r.postjobfree.com / Herndon, VA

• Seasoned Technology Professional with a proven track record of creating business value through Data & AI driven products.

• Recently transitioned to Data Science, Machine Learning and AWS-Cloud technology domain. Seeking a full-time opportunity

• Expert at stakeholder’s management, strategic planning, product design & management, analytics portfolio management, and transforming business objectives into data-driven solutions. TECHNICAL EXPERTISE

Deep Learning Architecture: TensorFlow, Keras, Pytorch, Artificial Neural Networks NLP: Text Analysis & Preprocessing – BERT, Genism, spaCy, Transformers, LSTM, RNN, Bidirectional LSTM RNN, Encoders, Decoders, Attention Models, Word Embedding, Word2Vec, AvgWord2Vec, TF-IDF, Bag Of Words, Unigrams, Bigrams, n-grams, Tokenization, Lemmatization, Stop Words Machine Learning Algorithms: Logistic Regression, SVM, Naïve Bayes, Random Forest, Decision Tree, KNN, K-Means, Isolation Forest, MCD. Statistics: Regression Analysis (Uni & Multi Variate), Lasso & Ridge Regression, Z-Score, Hypothesis Testing, Data Distributions (Gaussian, LogNormal, Poisson, Bernoulli &Binomial), Hypothesis Testing. Feature Selection & Engineering: PCA Standardization, Normalization, Pearson Correlation Coefficient, Spearman Rank Correlation, Chi-Square. Evaluation Metrics: Confusion Matrix, ROC / AUC Curve, F1-Score, Precision, Recall, Sensitivity, Specificity, MSE, MAE, RMSE, R-Square & Adjusted R Square, Gini Coefficient and Entropy

Tools: Python, R, PySpark, SQL, Tableau, AWS-Cloud – SageMaker, S3, Redshift, DynamoDB, Kinesis, Lambda, Glue, Athena, Quick Sight, Cloud Watch

PROFESSIONAL EXPERIENCE

Sr. Data Scientist Datalore Technologies Herndon, VA 08/2020 - Present Skills & Tools Applied: NLP, Machine Learning, SQL, Python, PySpark AWS-Cloud – SageMaker, S3, Redshift, DynamoDB

• Working with pharmaceutical clients to develop sales forecasting models and analyze the Covid-19 Impact on their sales.

• Developed Predictive and Forecasting models to analyze the fluctuation of drugs sales due to Covid-19 Pandemic.

• Applied NLP for performing text analysis based on number of hospitalizations cases and reports from reliable various resources.

• Performed text preprocessing with techniques – Word2Vec, Bag of Words, TF-IDF, Stemming, Lemmatization, Tokenization

• Implemented vectorization to transform the textual data into numerical data and used that as an input for the ML.

• Strategized the training and test data, evaluation metrics and model retraining, split the data accordingly.

• Applied various classifiers - Random forest, Decision Tree & Neural Networks and trained the model with 60% of the train data.

• Applied various optimization techniques to improve the model performance and achieved 84% accuracy.

• Evaluated the model accuracy using Confusion Matrix, Classification Report and Accuracy Score – F1 score, Recall and Precision.

• Strategic partnership with Product Manager, ML Engineer Data Engineers for intent gathering, designing, building data pipeline, productionizing, and operationalizing the model.

Sr. Data Scientist/Sr. Data Analyst Tekway Inc McLean, VA 10/2018 – 07/2020 Skills & Tools Applied: Machine Learning Models and Statistical Analysis: Regression Analysis (Uni, Multi Variate), Lasso and Ridge Regression, Classification Models: Naïve Bayes Classifiers, Logistic Regression, Random Forest, Decision Tree, KNN, Deep Learning Frameworks -Artificial Neural Networks, CNN, Keras. Tools: Python, PySpark, H2O, AWS-Cloud Worked with financial institution clients including Capital One Financial (10/2018-11/2019) Sentiment Analysis

• Developed Sentiment Analysis for Post Campaign Marketing customer feedback data using Natural Language Processing, ML and Deep Learning to predict customers behavior.

• Performed text analysis, text preprocessing, experimented with techniques – Word Embedding, Word2Vec, encoders, transformers.

• Applied algorithms in the Deep Learning Architecture – LSTM, Recurrent Neural Network, Bidirectional LSTM RNN and Encoders and Decoders and Attention Models

• Optimized the model performance by applying various hyper parameters and achieved the model accuracy of 81.2 %

• Evaluated the model performance using – F1-Score, Precision, Recall, Confusion Matrix Anomaly Detection

• Developed Anomaly Detection models for identifying unusual pattern of money transfer and fraudulent transactions using Pycaret.

• Performed data collection and integration, data cleaning, and exploratory analysis – data transformation, analyzing data distribution.

• and apply statistical methodologies to scale the data - Z-Score, MinMax, PCA, Remove multicollinearity

• Trained the model by applying algorithms – Isolation Forest (iforest), KNN, Minimum Covariance Determinant

• Evaluated the model performance using Confusion Matrices and model accuracy report.

• Achieved the model performance of 88%. Tested the model with the new set of data and the accuracy was as close as to the training accuracy.

Worked as Product Manager with Capital One Financial

• Supported the engineering effort to build “Data Scan Tool” for sensitive data discovery.

• Worked with stakeholders for intent gathering, negotiation and prioritization of business ask.

• Oversee the development team for a timely deliverable, product demo and productionizing the fully developed solution.

• Create user stories, epics and project updates in Jira and confluence. Executive presentation for product and project management teams

Data Scientist Intern National Institute of General Medical Sciences Bethesda, MD 02/2018 – 05/2018 Skills & Tools Applied: Deep Learning, Neural Networks, CNN, Radial Bias NN, TensorFlow, Caffe, PyTorch, Keras, Python, SQL, AWS

• Developed Image Classification Model for automatic detection of images of breast cancer and allow faster cancer diagnosis.

• Applied Back Propagation Neural Network (BPPN), Convolutional Neural Networks to build the image classification. and achieved the accuracy of 62.4%

• Optimized the performance of the classification model Radial Basis Neural Networks and achieved accuracy of 76.2%

• Experimented the Deep Learning models with various frame works like – TensorFlow, PyTorch, Caffe and Keras

• Extracted data from AWS-Cloud and on premises data stores. Performed data cleaning and exploratory analysis to derive hidden and meaningful insights.

• Chose relevant data that can add value to the classification model by data scaling, feature selection & feature engineering.

• Evaluated the performance of the models using – F1 Score, Precision, Recall, AUC-ROC

• Trained the model with various neurons, hidden layers, epochs, learning rate. Optimized the model by normalizing the data, Mini-Batch Gradient Descent, and with Momentum and Adam Optimization algorithms, transformation functions. Lead Technical Analyst – Data Management Xerox Business Services India 01/2008 – 07/2016 Expertise: Data Analysis, Analytics Engineering, Product and Project Management, Strategic Planning and Project Deliverables Prior to 2013 - Various team lead and individual contributor responsibilities across all technical areas of Analytics Engineering suite

• Oversea data engineering and analytics efforts across three global delivery centers, USA, Malaysia, and India

• Product and Project management for building Data Scan Application for sensitive data discovery and various other data products and create business value.

• Strategic partnership with stakeholders for intent gathering, negotiation, prioritize and translate business objectives into data driven solutions.

• Worked closely with data engineers to build data pipeline in AWS- Ecosystems for streaming and batch data, data transformation and data storage.

• Collaborated with Data Warehouse Architects for productionizing data into data lake, Software Engineers for developing APIs and UIs, Cyber Security and Data Management for Data Governance. Accountable for project design, development, testing and timely deliverables.

• Provided technical guidance, subject matter expertise for data analysis, statistical modeling, AB testing, Market-Mix models for financial marketing projects. Hands on experience building Predictive Models and Forecasting models using Python and R Programming Languages

o Regression models, Bayesian classifiers and Statistical Models, Logistic Regression, Random Forest, Decision Tree, K-NN, K-Means, Gaussian Mix Models, Time Series Models (ARMA, ARIMA, MA), ANOVA, Chi-Square, Z&T Test, Mann Whitney U Tests Business Impact:

Reduced data security cost by 4%, created business value by delivering end-to-end “Data Scan” Solution

Oversea engineering and analytics efforts across three global delivery centers, USA, Malaysia, and India Sr. Technical and Business Analyst Accenture Bangalore, India 2005 – 2007 Technical Support Executive Polaris Group Bangalore, India 2003 – 2005 Customer Support Executive Manipal Informatics Manipal, India 2002 - 2003 AWS – TRAINING

• AWS Certified Developer Associate Training from A Cloud Guru

• AWS Certified Big-Data – Specialty 2019 training from A Cloud Guru

• AWS Certified Machine Learning Specialty Training from A Cloud Guru EDUCATION

• M.S. Data Science The George Washington University, Washington, DC 2016 – 2018

• PGDCA Madurai Kamaraj University, Tamilnadu, India

• Bachelor of Sciences University of Mysore Karnataka, India

Contact this candidate