SUMMARY:

• Data scientist with experience building really fast and accurate machine - learning models in Python and R, also understand big data technology like Hadoop. Eager to work in a highly prudent environment where real skills and decisions are vital part of the job.
• Over 12+ years of total IT experience in analysis, design & development of business
• Over 4+ years of experience in ETL development, using Informatica Power center and Business Objects

TECHNICAL SKILLS:

Predictive Modeling Technique: Linear Regression, Logistic Regression, Segmentation and clustering, Decision Trees, Random Forest, Support Vector Machine and K Nearest Neighbor Classification, Feature Engineering

Statistical Methods: Regression models, hypothesis testing and confidence intervals, principal component analysis and dimensionality reduction.

Data Management skills: Reading Raw data files, Merging, Sorting, visualizing, Handling missing values, Handling programming errors, Appending of various datasets

Analytics software: R, SparkR, Python (scikit-learn, numpy, scipy, pandas), Hadoop(HDFS,Pig,Hive,HBase,Sqoop), SQL, Linux, Oracle

ETL Tools: Informatica Power Center (Informatica Designer, Workflow Manager, Work flow Monitor)

Business Intelligence: Tableau, Business Objects products such as Reporter, Designer, Webi, Oracle Discoverer/reports and Sales force Wave Analytics

PROFESSIONAL EXPERIENCE:

Data Scientist

Confidential

Responsibilities:

• Anomaly is something that deviates from what is standard, normal or expected. Worked on anomaly detection using algorithms like DBSCAN, Hierarchical clustering, Control Charts, Local Outlier Factor (LOF), studied patterns for high warranty claims.
• Anomalous warranty claims are selected for “Root Cause Analysis” and can result in prevention of escalation or warranty claims.

Data Scientist

Confidential

Responsibilities:

• Worked on Survival analysis model - used Kaplan-Meier Non parametric model for Univariate variable and Cox proportional hazard model for Multivariate analysis.
• We were interested in how long drafted Retirement Eligible employees stay at UPS, the duration of time leading Confidential to the event of interest is called Survival time. In this case the survival time is the number of months that a retirement eligible employee stayed with the company before retiring.

Data Scientist

Confidential

Responsibilities:

• Consumer complaint resolution is important to any business. If we are able to predict this, consumer likely disputed can be given more attention as to how the complaints are handled as well as how to convincingly convey the final conclusions to them.
• Designed and build production-ready machine-learning Logistic regression, Worked extensively on Feature Engineering, Exploratory Data Analysis. Handled categorical variables with more than 100 levels and created dummy variables for these successfully. Tried Support Vector Machine and K Nearest Neighbor Classification algorithms.

Data Scientist

Confidential

Responsibilities:

• Designed and build production-ready machine-learning Logistic regression, Ensemble model boosting and bagging for vehicle breakdown or not. Worked extensively on Exploratory Data Analysis, short listed statistically significant variables, used scatter plot to detect the correlation between the variables, converted categorical variables to dummy variables. Used dimension reduction technique such as PCA. Clustered vehicle as per geographic location for breakdown.
• Predict fuel flow rate of airplanes during different phases of a flight:
• Designed and build production-ready machine-learning multiple Linear regression and Random Forest models for different phases of a flight, worked on Feature Engineering, created dummy variable, removed some of the non-significant variables and selected statistically significant variables. Checked multicollinearity among model predictors with VIF analysis. Validated model using validation set got satisfactory result.

Technical Lead ( Business Objects, Tableau, Predictive Analytics)

Confidential

Responsibilities:

• At Wealth and Investment Management, Confidential management has been challenged with availability of metrics which display their IT Management Status, Risks and Staffing. IT Management Dashboard is an initiative which delivers all these combined. The data warehouse combines data from various different applications:
• Project Milestones/Schedules, Manpower Efforts/Cost, Financial Costs, Compliance Issues/Risks, etc. and combines them into a single dashboard giving a high level view of all the Projects status, resources assigned and various costs being incurred at given snapshot and over a period of time. The dashboard provides capability to compare the forecast of efforts/costs against actual. Business Objects was used to develop various reports like Allocation report, Resource report, Staffing Profile, DT-Milestones, Program Financial reports.

Sr. System Analyst

Confidential, NYC

Responsibilities:

• Continued to work on A2000 as well as other projects for Confidential LLC, after it takeover DMS technologies in 2003. As a team lead I was involved in Business requirement gathering, design, hands on coding and managing /coordinating development and implementation. During this tenure I was associated with following major projects.

Confidential

Responsibilities:

• This project was aimed to develop and implement a central data warehouse to stage the products data, sales data, customer data etc for various retail clients. This was basically an archiving project for past season products and related complete EDI transaction data. I participated in full data warehouse design, development, deployment. Designing, Development of Universe, also deployed Business Objects for reporting purposes. Business Objects was used to develop various reports like sales trends, sales performance, Regional sales, financial reports etc

Confidential

Responsibilities:

• The project involves maintenance and enhancements of the existing EDI 852 sales analysis system. Informatica and PL/SQL are used as ETL tools and Web-intelligence and Desktop Intelligence as a reporting/analytic tool. The system was designed to meet the analytic reporting requirements for the corporate. The raw 852 sales data is translated and stored in source oracle database that provides the basis for unlimited reporting possibilities making strategic decisions for the company.

Confidential, NY

Responsibilities:

• This Project involves development of a Sales Management and Report Tracking (SMART) Datamart that consolidates sales information in a common location. The Datamart contains transactional-level data (850 purchase order, 852 electronic Product Activity Data, Advance ship notice 856, 810 Invoices). It also provides reporting and analysis capability across multiple dimensions to users.

Oracle Developer / Production Support

Confidential

Responsibilities:

• Responsible for gathering Business Requirements, creation of Technical Design Documents and user Documents.
• Designed and developed reports using combination of Report 6i and XML publisher for modules like Purchasing and Order management.
• Developed various custom reports and customized standard Oracle reports like Order Status Report and Inventory, Allocation and Sales Analysis as per the business requirements.
• Developed reports in Discoverer 4i and created various business areas, user views, custom folders and joins for reporting.
• Created Workbooks based on the Custom Folders with multiple parameters and calculations using Discoverer Plus and Discoverer Desktop.
• Expertise in various oracle tools like Oracle Enterprise Manager (OEM), Toad etc.
• Used Microsoft SourceSafe as version control repository

Analyst/Programmer

Confidential, NYC

Responsibilities:

• I worked in this start Confidential as one of their first few core Programmers. I contributed passionately in development and enhancement of various modules of their Oracle based ERP package A2000. This software is designed and developed by DMS Technologies specifically as per needs of apparel Industry. Today A2000 have more than 200 plus implementations for clients in Apparel, Footwear, Jewelry, Fashion accessories and Home Furnishings.
• This ERP software helps organizations to build and implement systems that reduce cycle time related to internal decision-making, purchasing, accounts receivables and marketing through technology focusing on strategic areas such as merchandizing, marketing, procurement, store operations, and electronic commerce.

Environment: Oracle 7.3/ 8.0, PL/SQL, Toad, Developer 2000 Forms 4.5/5.0/ Reports 3.0, Windows 98/NT