Amitesh Kumar

• Collaborated with data scientists to develop a strategy for deploying ML and AI solutions at scale across an organization.
• Developing a predictive model to solve the business problem of Turnover based on my Prediction on Logistic Regression and Random Forest algorithms and creating an End-to-End Business Intelligence solutions delivery to optimize the ease of decision making
• Work with a Data Science Team Lead, data science management, business operations, and product management to assess the potential value and risks associated with business problems that have the potential to be solved using machine learning and AI techniques
• Application of various machine learning algorithms and statistical Modelings like decision trees, text analytics, natural language processing (NLP), supervised and unsupervised, regression models, social network analysis, neural networks, deep learning, SVM, and clustering to identify Volume using Scikit-learn package in python
• Collaborated with data engineers and operation team to implement ETL process, wrote and optimized SQL queries to perform data extraction to fit the analytical requirements
• Used Python (NumPy, SciPy, Pandas, Scikit-learn, Matplotlib, Seaborn) and Spark (Pyspark, MLlib) for model development for analytical purposes
• Improved the accuracy of the models using bagging and boosting techniques to minimize variance and improve predictive accuracy
• Perform Data Cleaning, features scaling, and features engineering using pandas and NumPy packages in Python.
• Developed predictive models using various machine learning algorithms and techniques such as linear regression, logistic regression, decision trees, support vector machines, neural networks, and ensemble methods.

• Worked on data processing on very large datasets that handle missing values, creating dummy variables and various noises in data
• Performed data pre-processing tasks like merging, sorting, finding outliers, missing value imputation, data normalization, and preparing it for statistical analysis
• Implemented ridge regression and subset selection methods to choose the most statistically significant variables for analysis
• Used machine learning algorithms such as Linear Regression, Ridge Regression, Lasso Regression, Elastic net regression, KNN, Decision TreeRegressor, SVM, Bagging Decision Trees, Random Forest, and XGBoost
• Built a classification machine learning model in Python to predict the probability of a customer defaulting in credit card payment and improved the model's accuracy
• Built compelling data stories and visualizations to influence decision makers.
• Used F-Score, AUC/ROC, Confusion Matrix, and RMSE to evaluate different model performances
• Performed data visualization and Designed dashboards with Tableau, and generated complex reports, including charts, summaries, and graphs, to interpret the findings to the team and stakeholders
• Predicted potential credit card defaulters with 84% accuracy with Random Forest
• Provide expertise and consultation regarding consumer and small business behavior score modeling issues and advise and guide the risk manager using the models in strategies
• Participate in strategically critical analytic initiatives around customer segmentation, channel preference, and targeting/propensity scoring
• Leverage a broad stack of technologies Python, Docker, AWS, Airflow, and Spark to reveal the insights hidden within huge volumes of numeric and textual data
• Deployed models as a Python package, as API for backend integration, and as services in a micro-services architecture with a Kubernetes orchestration layer for the dockers containers Work closely with business stakeholders, Financial Analysts, Data Engineers, and sound organizational decisions
• Work with Data Engineers to determine how to best source data, including identification of potential proxy data sources, and design business analytics solutions, considering current and future needs, infrastructure and security requirements, load frequencies, etc.

• Utilized Agile Scrum Methodology to help manage and organize a team of 4 developers with regular code review sessions
• Responsible for conducting data-driven strategic analyses and developing internal decision-making tools/products to enable retail team decision making
• Analyzed user requirements, designed and developed ETL processes to load enterprise data into the Data Warehouse.
• Written multiple MapReduce programs to power data for extraction, transformation, and aggregation from multiple file formats, including XML, JSON, CSV & other compressed file formats
• Developed a predictive model using historical and current data to identify interested customers for Email Campaigns
• Participated in feature engineering, such as feature intersection generating, feature normalization, and label encoding with Scikit-learn pre-processing
• Used Python (NumPy, Scipy, Pandas, Scikit-Learn, Seaborn) and Spark 2.0 (PySpark, MLlib) to develop various analytical models and algorithms
• Utilized Spark, Scala, Hadoop, HBase, Kafka, Spark Streaming, MLlib, Python, and various machine learning methods, including classifications, regressions, dimensionally reduction, etc
• Implemented, tuned, and tested the model on AWS EC2 to get the best algorithm and parameters
• Participated in all phases of data mining; data collection, data cleaning, developing models, validation, visualization, and performed Gap analysis
• Collected data needs and requirements by Interacting with the other departments.

· Managed Peoplesoft upgrade project for the retail chain, which saved the client $2M in license fees and improved employee experience driving 100% adoption of a new system.

· Assisted client through the RFP phase by interviewing stakeholders and documenting precise requirements; assisted with cost and value analysis of different vendors and built use case for vendor selection.

• Led a team of developers in the successful implementation of a PeopleSoft Financials module for a financial services company, resulting in improved financial reporting and a 15% reduction in manual data entry
• Developed and implemented a custom integration solution between PeopleSoft and an external time and attendance system for a healthcare organization, resulting in improved accuracy of employee time tracking and a 15% reduction in payroll errors
• Led a team of developers in the successful implementation of a custom compensation module in PeopleSoft for a telecommunications company, resulting in improved accuracy and efficiency of the compensation process and a 15% reduction in compensation processing time
• Designed and developed a custom payroll module in PeopleSoft for a healthcare organization, resulting in improved accuracy and efficiency of the payroll process and a 20% reduction in payroll processing time
• Peoplesoft and Python Developer.