SUMMARY:

• 13+ years of IT experience focused on Big Data & Java/J2EE/JEE
• 2.5 years of Big Data experience in Hadoop, Machine learning, Map Reduce, Hive, Pig
• 10 years + extensive experience on Java, J2EE, Spring, Struts, Design Patterns
• Strong experience on Weblogic, JBOSS, TOMCAT

TECHNICAL SKILLS:

Big Data Technologies: Hadoop, Map Reduce, Hive, Pig, Mahout, Kafka, Storm

NoSQL databases: MongoDB

Languages: Core Java - Implementing object oriented principles using Abstraction, Encapsulation, Inheritance and Polymorphism with access modifiers and specifies, extensively used Collections, Java IO, RMI, Multithreading, JDBC and etc. J2EE - Servlets, JSP, EJB, JMS, JNDI, Java Mail, parsing XML using SAX and DOM parsers. Written and deployed applications for Java/J2EE/JEE on Windows and UNIX platforms.

Others: Hadoop, R, Hive, Pig, Mahout, Spring Batch, Kafka, Storm, Avro, Octave and Apache Ant

JAVA Versions Worked: 1.2, 1.3, 1.4, 1.5 and 1.6

AJAX: Written AJAX code using Java Script, XML,Servlets and JSP. Extensively used DWR framework to develop AJAX based GUIs' and Jquery

Database servers: Oracle 7, 8, 9 and 10, MySQL 5, PostGres

Design/ Modeling languages: Designed and developed UML - Class, Package, Object, Use Case, Sequence, Collaboration, State Chart, Activity, Component and Deployment diagrams using MS-Visio, ArgoUML editors.

Design Patterns: Singleton, Factory, Abstract Factory, Value Object, Front Controller, MVC, DAO, Service Locator, Business Delegate and Session Fa ade.

Web/App Servers: BEA WEBLOGIC 9/8.1/7.0, Tomcat 5.x, JBoss.

Design Tools: Microsoft Visio 2000, Visual paradigm, Altova UML

Java Frameworks: JUnit, Hadoop, Mahout, Struts 1.2.x, spring 2.5 and Hibernate 3.0

Web technologies: HTML, JAVASCRIPT, XML and CSS

IDE: Eclipse 3.0, JCreator and NetBeans

Operating Systems: Win 98/NT/2000/XP, UNIX, Solaris and Red Hat Linux

Version Control System: Rational Clear Case, Win CVS, SVN and CVS (built in for Eclipse and command line)

Optimization: Optimized code by different layers at web applications and at Core Program and DB layer at standalone applications level. Optimized multi-threaded applications by analyzing Thread Dumps. Optimized Hive, Map Reduce programs

PROFESSIONAL EXPERIENCE:

Confidential, Denver, CO

Big Data Developer/ Architect

Responsibilities:

• Design of the Knewton Project
• Coding of the Knewton Project Map Reduce modules.
• Managed the Knewton project by leading a team of 3 people.
• Integrating the work flow of the Knewton projects with Spring Batch
• Writing the component test cases for Random Forest, Spring Batch job reports of Confidential .
• I was part of the interview panel to select new lateral joinees

Environment: Java, Map-Reduce in Java, Hive, Spring-Batch, CentOS linux

Confidential, Phoenix, AZ

Sr Big Data Developer/ Technical Architect

Responsibilities:

• Consumer Relevant Project is to categorize the Merchants to a particular Industry codes.
• The first stage of categorization was achieved with Text Mining, a Map Reduce program that would do a 1-gram match, 2-gram match and a 3-gram match against a mapping matching file for obtaining an appropriate industry code.
• The second stage of categorization is done using Mahout (Random Forest). The output of Text Mining, Random Forest form the input to the arbitration logic along with the SIC codes (Suggested Industry Codes).
• The output of the arbitration logic gives the top 2 industry codes for all the merchants.
• The complete workflow is automated with well-defined Spring Batch tasklets.

Environment: Java, Map-Reduce in Java, Mahout, Hive, Spring-Batch, Red hat linux, 40 node cluster

Confidential

Responsibilities:

• Confidential is an ETL tool to generate queries that can be used to extract data based on a Meta data input.
• The Confidential Project has a query builder module that is built upon the SQL ORM to generate optimized Hive queries.
• The query builder module gives the flexibility for sampling and filtered result.
• The query execution module executes the queries generated by the query builder and dumps the data and the data dictionary in a HDFS location to be consumed by the downstream systems.
• The complete workflow is automated with well-defined Spring Batch tasklets.

Environment: Java, Spring-Batch, JAXB, Hive, Red hat linux, 40 node cluster

Confidential

• GoupOn POC is a tool to download GroupON coupons using rest API and store it MongoDB. Using Hive we do data transformation to modify the data.
• Then we have apply Jaccard Coefficient to find similiarity between the existing merchants and GroupOn merchants, this is important step as we could then Confidential the deals using the existing scoring logic to rank the offers.
• This involved giving presentations and demos to the senior management team of the client
• Successfully delivered on multiple initiatives on big data analytics and large data processing using Hadoop M/R, Hive and Mahout.
• Implemented K-Means clustering using Octave for a machine learning use case
• Implemented Linear Regression using Octave and R for a machine learning use case
• Implemented similarity comparison using Jaccard co-efficient.
• Took various sessions on writing effective Map reduce jobs and executing them over Large Hadoop clusters.
• Writing optimized Pig scripts for data analysis - using Pig Latin extensively for analyzing semi structured data such as log files and huge data sets stored in HDFS.
• Random Forest implementation using Mahout and R
• Lead a team of 8 people, was primarily involved in the project. Additionally was responsible to mentor people on Hadoop, Hive, Pig and Machine Learning algoithms. I was also the R consultant for the team to bridge the communication between Data scientists and the Big Data development team
• Design of the consumer relevance project, Confidential Project
• Coding of the Random Forest module for the Consumer Relevance project, Spring Batch work flow for both Consumer Relevance project and Confidential, Serializing the result of each step of Confidential using JAXB.
• Prepared visio process diagrams for each process in Consumer Relevance and Confidential .
• Reviewing the Hive scripts and optimizing Hive for performance.
• Prepared high level design and low level design documents.
• Analyzed, designed and developed the Random Forest module.
• Optimized the text mining module by designing and being a part of the team that wrote custom map reduce programs which reduced the running of the text mining module from 6 hours to 15 minutes.
• Managed the consumer relevance project by leading a team of 8 people.
• Managed the Confidential project by leading a team of 5 people
• Integrating the work flow of both the Confidential and Consumer Relevance projects with Spring Batch
• Writing the component test cases for Random Forest, Spring Batch job reports of Confidential .
• Review of the component test cases of the team in the Consumer Relevance and Confidential projects
• Design, coding and demos of POCs like GroupON to the customer.
• Mentoring new joinees on Big Data and Java
• Single point of Contact from offshore for the projects Consumer Relevance and Confidential
• Interact with Data scientists and help the team to understand R scripts to run the use cases using Map Reduce
• Technical Presentations in the Big Data Architect forum, I have given 2 presentations namely Introduction to R and Introduction to Phoenix, as SQL wrapper for HBase.
• I was part of the interview panel to select new lateral joinees

Confidential

Technical Lead

Responsibilities:

• Owned the calculates invoice amounts and accrual amounts, generates invoices and accruals. RMS system is also used for generating extensive reports such as revenue forecasting, revenue variance and operational reports.
• Worked on the re-design and implementation of a more reliable exception-handling framework across the EJB layers.
• Worked on a multi-threaded client to generate accruals. The accrual generation client was used to simulate a complex production scenario and solve a synchronization problem.
• Lead the Development team of 10 members from offshore. I was responsible for managing the deliverables, interacting with the Business team to understand the requirements
• Design, Coding, Unit testing of the Revenue Forecasting module exception framework
• Design and coding of a multithreaded test client to solve a concurrency issue leading to wrong calculation if Daily Fee Caluclation module
• Solved critical issues of Invoicing calculation in the Daily fee calcualtion process and a problem idempotent requests
• Planning and support the Billing analysts during the billing cycles to enusre proper invoicing
• Performance tuning of the Java modules and the Oracle modules
• Interacting with the Buisness team in understanding and priortizing the requirements
• Managing the Revenue management system IT team which consisted of 8 members
• Planning, Design and requirement gathering of the Revenue management deliverables.
• Mentoring the team on technical aspects
• Technical consultant for the team for helping them understand EJBs, JSTL, debugging calculation issues for invoices and accruals
• Project planning using Microsoft project plan and allocation of work to the team
• Review of the team code and changes.
• Appraisals of the team with clear cut goals and improvement plans if any

Environment: Weblogic on solaris, Oracle 9i, EJBs, JMS, Ant, Crystal Reports.

Confidential

Tech Lead

Responsibilities:

• Operation Risk Inventory or ORI is an inventory with Operation Risks and its corresponding action plans/dispensations.
• Heatmap is a reporting tool linked with ORI.
• APT is a tool which helps in tracking the action plans flagged for tracking. ORI uses struts and Heatmap and APTT uses Spring as its MVC frame work.
• The database access is through hibernate which is a part of the spring framework.

Environment: Tomcat on windows XP, Java, Spring MVC, Hibernate, Ant, SVN, TortiseCVS, UML, Ajax, Tomcat, POI, DWR, Oracle 9i, Windows XP

Client: Confidential (Worked onsite at London)

Mobile Number Portability UK:

Responsibilities:

• Mobile Number Portability allows a customer to change his operator to another operator while retaining his mobile number or MSISDN.

Environment: Weblogic 8.1 on Solaris and windows XP, JMS, Enterprise Java Beans(EJB), J2EE, WebLogic, Java, Ant, CVS, TortiseCVS, UML, Velocity templates, Oracle 9i, Solaris and windows XP

Confidential

Batch Manager

Responsibilities:

• Batman was to automate the pre-provisioning mechanism, so as to reduce improving the efificiency.

Environment: Weblogic 8.1 on Solaris and windows XP, JSP, Servlets, Oracle, Jakarta-Struts, EJBs

Confidential

Responsibilities:

• Mobile Number Portability allowed a customer to change his operator to another operator while retaining his mobile number or MSISDN.
• There was a central repository for Ireland called the National Porting Database for porting.
• This was vendor independent and kept track of all relevant information and other related information for porting.

Environment: Weblogic 8.1 on Solaris and windows XP, JSP, Servlets, Oracle, Jakarta-Struts, EJBs

Confidential

Technical Lead

Responsibilities:

• Project planning and tracking using microsoft project plan
• Heatmap is the reporting tool for Operation Risk Inventory in UBS. Heatmap provides access to risk data based on the individual role and the corresponding division in the organization. Assessment of risks in Heatmap was initially designed for individual node groups. Re-designed and implemented the tool for managing assessments across multiple node groups.
• Design and coding of XML parsing using XPath.
• Interacting with the Client IT team to have discussion on Requirements and Design
• Manage the SIT and UAT cycles from the development team perspective
• Worked on the design and development of Mobile Number Portability solution using JMS, session beans, entity beans with container managed persistence.
• Interacting with multiple vendors and customer during my tenure of the onsite lead
• Automated the Mobile Number Portability (MNP) deployment with an Ant deploy script for packaging and configuring property files. The Ant script was responsible for modifying the BEA WebLogic system configuration by creating JMS queues, data sources and deploying the JAR file. The automated deployment helped reduce the errors and time involved in setting up the MNP application in Test and Production environments
• Automated MNP test cases with comprehensive unit test cases that helped reduce the testing time from 3 days to 0.5 day.
• Involved in the migration of the MNP application from BEA WebLogic 6.1 to 8.0
• Authored a white paper on MNP UK and Black listing of Mobile Phones in UK.
• I worked as an onsite lead in London for 2+ years, managing the deliverables and co-ordinating with the offshore team. I worked for 1 year as the delivery lead managing a team of 15 people and the deliverables from offshore

Confidential

Senior Software Engineer

Environment: Tomcat on NT, JSP, Servlets, Oracle, Jakarta-Struts

Responsibilities:

• Responsible for the design and development of the Booking Centre module.
• XML file download of the trading product information with well-defined XML schema.
• Worked on the generation of shipping documents for a Trade Express application.
• Coding, Unit testing of defects and change requests

Confidential

Responsibilities:

• J2EE internet Banking Site.
• This J2EE application was implemented for a company sponsored training class; the application provided customers with Internet banking services.

Environment: EJB 1.1,JSP, JBoss

Database: postgresSQL

Confidential

Responsibilities:

• Developed an application that was used to track various project details.
• The tracking system was designed with three kinds of users in mind: An administrator, a project manager and clients.

Environment: Orion on Windows 98, JSP, Servlets, PostgresSQL

Confidential

Sotware Engineer

Responsibilities:

• Worked with requirement gathering, design, coding and testing of the Web performance-testing tool. This tool was an extension of the open source JMeter web-testing tool, with enhanced customized reports specific to the client requirements.
• Developed an applet-servlet chat to facilitate communication between the consultants at Labyrinth and its clients.
• Authored a white paper on Knowledge Management.
• Customer interaction to understand the requirements for various tools and solutions
• Deployment of our products in the client systems
• Design, coding and Testing of web based applications; f Swing based standalone products for customers; applet-servlet chat to ensure that only the last messages get appended without refreshing the whole chat window; Java based solution to migrate the existing ERP solution to a java based solution

Environment: Apache server on windows, CGI-Perl, Flat files for data storage