Candidates About

 

Debu Sinha

About me

·         5+ years of experience in software development. 

·         4+ years of experience in Big Data development.

·         Extensive experience in working directly with C-suite and

understanting business use case while providing end to end solutions.

·         Extensive experience in using Cloudera Hadoop ecosystem.

·         Ability to be highly productive while working in a team or as an individual contributor.

 

Experience

 

Throtle Inc. (Previously V12 Group), Red Bank, NJ                                                                  Nov 2016 – present

Digital Ad-Tech, Data Science Team

Sr Software Dev – Data Scientist and Team Lead

·         Use Spark & Hive to run/schedule ETL jobs for generating reports along with performing traffic pattern and user behavior analysis using machine learning.

·         Developed a predictive model using Spark MLlib that flags the traffic from cookie clearing and cookie blocking browsers using supervised machine learning.

·         Architected, implemented and migrated Throtles core client onboarding platform to AWS from the bare-metal infrastructure in Softlayer.

·         Architected the VPC and subnets from scratch while using multiple AWS services while working with one system admin and one DevOps resource.

·         The core client-onboarding infrastructure adds approximately 70M targetable devices (cookies + mobile devices) monthly into Throtles ID graph and supports 1.5 Billion monthly pixel match and sync events.

·         Leading the Data Processing team consisting of 3 Hadoop developers and 1 Data analyst for designing the data curation strategy while working directly with CTO and VP of Engineering.

·         Regularly perform training for account managers and sales team to educate them about the onboarding product features for client engagement.

·         Drive meetings with clients and integration partners.

 

V12 Data (Previously V12 Group), Red Bank, NJ                                                                  Oct 2014 – Nov 2016

Digital Ad-Tech, Data Science Team

Sr Software Dev – Data Scientist and Team Lead

·         Architected and implemented V12 Groups core client on-boarding platform.

·         Architected and Performed capacity planning for setting up an Enterprise Data Hub/Data Management Platform(DMP) capable of handling Billions of monthly transactions.

·         Deployed and managed Cloudera 5.4 Hadoop(YARN) cluster with an active capacity of 500TB over 22 servers.

·         Architected and implemented the data processing pipeline using Kafka 0.8.1.1 and Flume1.5, Zookeeper 3.4, Hive and Sqoop2.

·         Architected and implemented an in-house real-time text search and visualization utility using Logstash, Elastic Search 1.7 and Kibana 4.

·         Created custom Kafka producers for handling accounting logs using Java from 9 geographical distributed PMTA servers generating close to 100 million unique events per month.

·         Lead a team of 2 Linux system admins and three software developers to support the Enterprise Architecture and monitoring.

·         Architected and implemented an auto-scaling Selenium Grid 2 Test automation suite over AWS.

·         Architected and implemented an in-house pixel-server for performing cookie syncing between end user browser and DMP for engagement customer tracking and segment building.

·         Created scripts using Hive and Linux Shell scripting to run and monitor Map-Reduce jobs.

·         Currently working on adding real-time data processing capabilities in the Enterprise Data Hub.

·         Working directly with CEO and CTO to drive the future of this project and driving the discussions about how the data collected in the Enterprise Data Hub can be used and mined using predictive analytics and machine learning to get valuable information.

·         Developing hands-on video tutorial for people who want to learn about Big Data, machine learning, and data science in spare time along with sharpening my skills and learning something new every day.

Bank of America, Merrill Lynch, New York, NY                                                                       Feb 2013 – Sep 2014

Global Market Risk Technology Operations (GMRTO) Sales Portal Team

Officer – Technology Analyst and Developer (DevOps Hadoop developer)

I was an active member of Hadoop user group within Bank of America which stretched across multiple LOB's where we shared our proof of concept ideas and results.

·         Implemented a data collection framework using Flume and Kafka to ingest TB's of access logs each day into HDFS.

·         Performed statistical analysis of user access information and developed a model to detect and report potential robotic users.

·         Used clustering algorithms to segment users into different segments for our sales team using Mahout.

·         Designed & Implemented a multi-node Hadoop Cluster for R&D purposes with Hive, HBase, YARN, and Zookeeper.

·         Developed an internal automated testing framework using JUnit 4 in the cloud using services from SauceLabs.

·         Delivered a recommendation system for sales team based on Hadoop architecture while working with Analytics Data Advisory team.

·         Analyzed weblogs with Datameer from different LOB’s and managed to draw a correlation of how our clients react to the market news.

·         Developed a system for our sales team that produces multiple lists of potential customers who may take an interest in various existing products based on current market news.

 

Bank of America, Merrill Lynch, New York, NY                                                                      Jun 2012 – Aug 2012

GMRTO Sales Portal Team

Summer Intern – Technology Analyst and Developer

·         Supported the integration of Oracle Golden Gate into the sales portal servers which provided scope for hot-hot data replication.

·         Built up a framework for automating Selenium test case runs on cloud infrastructure. The work brought down testing costs by more than 60%.

·         Formulated an algorithm to compare two very large database schemas consisting on BLOBs and CLOBs. The implementation of this algorithm reduced our database consistency checking time down from 3+ days to about 4 hours.

·         Performed a POC to compare the benefits of PhantomJS to that of Selenium according to our test cases

·         Assisted various lines of business to migrate to sales portal by developing modules in ExtJS and Liferay.

·         Developed dashboards in Splunk to analyze user actions from server access logs.

 

Johns Hopkins University, Baltimore, MD                                                                                   Jul 2011– Jan 2013

Blackboard Development and Support Team

Student Programmer – Information Technology

·         Developed two new LoginAs building block features, one for logging admin activity and the other for viewing a specific user’s Blackboard interface

·         Analyzed the student and faculty database of JHU and generated user reports that allowed for meaningful suggestions to enhance the quality of Blackboard for all users.

·         Created Course Life building block to facilitate course archiving feature for JHU.

·         Contributed these modules to the open source community.

 

Accenture Services, Chennai, India                                                                                           Jan 2011 – Jun 2011

Health & Public Sector Domain

Software Engineer- Associate

·         Worked with Stanford University Hospital to migrate data from OLTP (database) to data warehouse performing ETL operations using Informatica.

·         Collaborated with on-site teams and business analyst to translate business requirements into high-level technical design documents.

·         Developed OLAP cubes for hospital patient historical data to facilitate reporting needs of BO team.

 

Certifications

·         Cloudera Certified Developer for Apache Hadoop (CCD - 410)

·         Data Science Specialization Certificate (Johns Hopkins University/Coursera) pursuing

·         Red Hat Certified Engineer

·         IBM certified Rational Application Developer with WAS v 6.1

·         Certified Database Associate - IBM DB2 (000-730)

·         AWS Certified Developer Associate

 

Skills

·         Languages: Java, J2EE, Python, C, C++, HTML, R

·         Technologies: Hadoop, Core Spark, Hive, Pig, Sqoop, Oozie, Zookeeper, Flume, Amazon EC2, Kafka, Avro, Elastic Search, Kibana, Logstash

·         Hadoop Distributions: Cloudera, Hortonworks

·         Servers: Apache, Tomcat, WebSphere Application Server, JBoss, Tornado

·         Development Tools: Eclipse, ArgoUML, phpMyAdmin, IBM RAD, Informatica

·         Machine Learning.

·         Relational Databases: IBM DB2, MSSQL, MySQL, Oracle

·         NoSQL Databases: CouchDB, MongoDB, Cassandra, HBase

·         Continuous Integration Tools: Jenkins

·         Testing frameworks: SauceLabs, Selenium, PhantomJS, JMeter

·         Version control: Git, TFS

·         Operating System: Windows XP, Windows 7, Red Hat Enterprise Linux, Centos, Ubuntu, Mac OS

 

Education

Johns Hopkins University, Baltimore, MD and Coursera                                                            Apr 2014 – Present

·         Data Science Specialization Certificate (9 Courses 1 Capstone project)                                           

Johns Hopkins University, Baltimore, MD                                                                                 Jul 2011 –Feb 2013

·         Master of Science in Computer Science with Graduate Research, 3.5 GPA

Rajiv Gandhi Technical University, Bhopal, India                                                                  June 2006 – Jun 2010

·         Bachelor of Engineering in Computer Science, Top 1% of class, 3.9 GPA

 

Academic Projects/ Competitions

Johns Hopkins University, Baltimore, MD                                                                                Dec 2011 – Jun 2012

Project Name – English-Hindi word alignment improvement using syntax 

Master’s Research – Machine Translation course

·         Worked on a research project to study the effect of syntax based translation on the quality of alignments produced by any standard aligner like Berkeley Aligner for low resource language Hindi.

·         Derived a new metric called as Diagonal Error Rate to study the relation between the syntax based transformations and their effect on alignment quality.

·         Documented the results and presented the findings in Master thesis.

 

Johns Hopkins University, Baltimore, MD                                                                                 Jan 2012 – Jun 2012

Project Name – Benchmarking performance of Greenplum (Massively Parallel Processing database), Cassandra and HBase (NoSQL databases) with social media Big Data set.

Programmer – Big Data

·         Compared and benchmarked these three databases on performance, I/O operations, throughput and scalability.

·         Set up individual multimode clusters for these three databases.

·         Perform ETL operations for each database.

·         Presented the findings in a class presentation.

 

Johns Hopkins University, Baltimore, MD                                                                               Oct 2011 – Nov 2011

Project Name – Dbox Online Resource Management System based on REST architecture

Programmer – Computer Networks

·         Led team of three developers to design and develop key modules performing both client and server side operations, like exchanging data over HTTP protocol using XML message passing.

 

Johns Hopkins University, Baltimore, MD                                                                                Jun 2011 – Oct 2011

Project Name – VNS Router and Firewall Implementation

Programmer – Computer Networks

·         Led a team of three to implement a routers functionality

·         Developed key modules to handle complete firewall features and TCP and UDP packets

·         Finalized the project infrastructure and put to use ARP cache feature

 

Truba Institute of Engineering in collaboration with IBM

Project Name –Stocks and Portfolio Management System                                                       Mar 2010 – Jul 2010

Developer – Major Project

·         Led team of four developers to perform requirement analysis and system design using various UML techniques (use case diagrams, sequence diagram, class diagram, collaboration diagram and deployment diagram).

·         Finalized project development infrastructures (source tree, coding-conventions, version control, and reporting/charting components), test plans and deployment plans.

·         Designed and developed key modules for the applications using JSP, JavaScript, AJAX and tag libraries.

 

IBM The Great Mind Challenge, Bhopal, India                                                                        Mar 2009 – Jul 2009

Project Name – Virtual Classroom Management System                            

Developer – National Level Coding Competition

·         Led team of four developers through the entire lifecycle and iterations of this project from concept to deployment and maintenance.

·         Analyzed requirements and system design using various UML techniques (use case diagram, sequence diagram, class diagram, collaboration diagram and deployment).

·         Finalized the project development infrastructures (source tree, coding-conventions, version control).

·         Designed and developed key modules for the applications using JSP, JavaScript, and AJAX.

·         Earned certificate of appreciation by IBM for landing into top 20 teams out of 14,000 in The Great Mind Challenge, project was used to start IBM’s internal online collaboration

 

 

Saraswati Vihar, local school in Bhopal, India                                                                        May 2006 – Dec 2006

Project Name: Online Library Management System

Developer – Volunteer

·         Gathered the requirements for an online library management system

·         Developed all supporting server side code and pages using MVC architecture

·         Tested the quality using JUnit

 

Publications

·         Exploring the Prospect of Secure Biometric Cryptosystem using RSA for Blind Authentication, International Journal of Wisdom Based Computing, Singapore. Volume 1 Issue 2 – Page 24-27
2011

 

·         Secure Biometric Cryptosystem for Distributed System, International Journal of Communication Networks and Security, Volume 1 Issue 2 – Page 27
Best paper award at ICCSI-2011, International Conference Bhubaneswar, India
2011

 

·         Exploring the future of Secured Biometric Data Authentication protocols using ECC and Blind Authentication, National conference on CSI, MANIT Bhopal.

2011