SUMMARY

• Lead Data Scientist with 4+ years of experience in solving business challenges using data analytics.
• Proficient in predictive modeling, data processing, data mining, and supervised and unsupervised algorithms for identifying patterns and extracting valuable insights for stakeholders.
• Highly skilled in Machine Learning and Deep Learning, and in creating interactive Data Visualizations.
• Possess solid understanding of Mathematics, Probability, and Statistics.

TECHNICAL SKILLS

Technical Skills: R, Python, SQL, SAS, SPSS, Stata, ForecastX, Minitab, Matlab, GitHub, Excel, PowerPoint, R Markdown

Visualization Tools: Power BI, Tableau, ggplot2, Leaflet, Shiny, Seaborn, Matplotlib

Machine Learning Algorithms: Linear and Logistic Regression, Polynomial Regression, Random Forest, K - Means, Artificial Neural Networks, Decision Trees, Naïve Bayes, SVM, K-Nearest Neighbor, Autoencoders, RBMs, Probit, PCA

Theoretical Assets: Multivariate Calculus, Linear Algebra, Parametric and Nonparametric Statistics, Probability and Inference, Econometrics, Time Series Analysis, Forecasting

Soft Skills: Strong Leadership and Interpersonal Skills, Excellent Verbal and Written Skills, Robust Business Acumen

Day-to-Day Skills: Data Extraction, Data Wrangling, Data Analysis, Data Visualization, Research

PROFESSIONAL EXPERIENCE

Confidential, Farmington Hills, MI

Lead Data Scientist

Responsibilities:

• Currently, using ForecastX and R to conduct Time Series Analysis to forecast off-lease vehicle requests to help the dispatch department create better inspector schedules. Models used include Exponential Smoothing Models, Time Series Decomposition model, ARIMA models, and Artificial Neural Networks
• Built interactive visualizations (Shiny app) for the dispatch team that display inspectors’ work area coverage using Leaflet, and Shiny packages in R and Power BI
• Utilized SQL to compile data from different databases, R for data wrangling and statistical analysis, and R Markdown to turn analysis into high quality reports, dashboards, and presentations
• Used supervised learning algorithms like SVM, Naïve Bayes, K-Nearest Neighbor, and logistic regression to solve classification problems with high accuracy. Used concepts like cross-validation method and the confusion matrix
• Utilized Python’s libraries Pandas for data wrangling, Matplotlib and Seaborn for data visualization, and Scikit Learn for data modeling
• Supervised the work of a new data scientist in stages of the data science process such as data extraction, data munging, data analysis (K- means clustering and Box Jenkins models), data visualization, and result presentation

Predictive Analytics Engineer

Confidential

Responsibilities:

• Built two mathematical models to help the dispatch and the inspections teams determine their optimal staffing needs. Used an Erlang C model to predict call center load relying on Power BI and R to conduct the analysis.
• Created visualizations for AiM and Manheim using the ggplot2 and caret packages in R to communicate ideas to upper management that can aid in making data-driven decisions
• Used deep learning by designing and implementing different neural network architectures to predict the count and (Group, Area, Type, Location, and Severity) of damages on off-lease and auction vehicles using R packages like h2o, deepnet, neuralnet, and darch. Used deep neural networks with Restricted Boltzmann Machines, Autoencoders, and Anomaly detection
• Used machine learning algorithms to perform predictive analytics to help AiM make the vehicle inspection process more efficient. Analysis affected training of inspectors and the inspection app design. Models utilized included Linear and non-linear regression, Logistic Regression, Multi-logistic Regression, Random Forests, Decision Trees, and Naïve Bayes
• Applied concepts of probability, distributions, and statistical inference on data sets to unearth findings using comparisons. Used t-tests, F-tests, P-values, Bayes’ Theorem, hypothesis testing, Pearson and Spearman Correlations, ANOVA, Adjusted R-squared, AIC and BIC

Statistical Researcher

Confidential

Responsibilities:

• Used Stata and SAS to conduct descriptive and inferential statistics to analyze arbitrations occurring at Manheim auctions and build inferential models to understand the relationship between arbitrations and different independent variables like location, sales, and managers
• Use R to create visualizations to tell data stories to upper management. Visualizations created included Line plots, histograms, bar graphs, pie charts, and boxplots
• Fit various models like Probit and linear regression to understand the relationships between vehicle damage count on one hand and mileage, make, model, color, and model year on Ford two-year lease vehicles on another
• Operated in an agile environment with daily scrum meetings; standup meetings, burn-down charts, presentations and reviews

Confidential

Research Assistant

Responsibilities:

• Researched the factors that influence the sensitivity of customers to unethical & environmentally unfriendly behavior of companies to thoroughly understand consumer behavior.
• Compiled relevant journal articles, wrote a literature review, and performed data analysis using SAS to perform high-quality research. Created visualizations in Tableau

Confidential

Adjunct Lecturer

Responsibilities:

• Taught seventeen upper and lower division sections of the following five economics courses: Microeconomics, Macroeconomics, Intermediate Microeconomics, Managerial Economics, Public Finance and Fiscal Policy
• Produced syllabi, led class instructions and discussions, and evaluated and advised students

Confidential

Adjunct Lecturer

Responsibilities:

• Taught eleven sections of the following economics, mathematics, and statistics courses: Mathematics of Finance, Business Statistics, Business Mathematics and Statistics for Engineers, Freshman Calculus for Arts Students, Calculus I for Freshman Science, Principles of Microeconomics, Principles of Macroeconomics
• Taught students Minitab and SPSS in Statistics courses