Ashish Kumar Singh Email & Phone Number

San Francisco Bay Area

San Jose, California, United States

Bengaluru Area, India

Pune Area, India

Responsibilites: Work on building scalable , efficient and reliable data reporting at PubMaticAudience Reporting : - Responsible for Audience Segment Reporting: included distinct users and revenues for publishers - Used Probabilistic Data Structures (HyperLogLog) for cardinality estimation - Revamped end-to-end audience data reporting- Wrote global scheduler for Multi DC dependent workflows- Wrote functions in Apache Phoenix for HBase to aggregate Hyperloglog masks in real time Proof of Concepts: - Compared Hadoop and Spark for 3 use cases : cardinalty estimation , multi stage workflows , grouping set generation for cubes- Tried Elastic search for matching and finding combination of dimensions for reporting Tools and Technolgies : - Programming languages : Java and Scala - Big Data execution engines : Hadoop , Spark - NoSql Databases: HBase - Scripting : Perl / Shell Talks and Seminars : - Speaker at Spark Summit 2015 on "Migrating Complex data aggregations from Hadoop to Spark "​- Attended Fifth Elephant '14 conference on Big Data in Bangalore

Pune Area, India

Title: Healthcare AnalyticsDuration: from Dec 2011 to PresentPosition: Hadoop/Mapreduce DeveloperEnvironment: Red Hat Linux Technologies & Databases: Hadoop , Pig, Hive, Sqoop, HBase ,DB2 , Java .Description:Top US Healthcare firm wanted to utilize data obtained from millions of claims everyday for Fraud detection and patient analysis. This project focuses on big using data analytics for greater research into healthcare data .The Fraud detection algorithm developed is constantly updated and improved while patient data is used to provide greater insights to Medical Practitioners . DB2 connectors and COBOL flat files are used as input . Responsibility:● Writing native Mapreduce programs in Java● Developing Pig Script with UDFs in Java ● Testing Mapreduce programs using JUnit● Improving Fraud detection and Patient history analytics ● Applying data mining algorithms using Mapreduce● Transfer data using DB2 connectors and COBOL flat filesPROJECT -2Title: Major Retail ProjectDuration: from May 2011- Nov 2011 Position: Java developerEnvironment: window XPTechnologies & Database: J2EE, XML, DB2.Description:This project involves B2B (e-commerce) application including marketing, sales, customer and order processing functionality and automating the Quotes and orders between the organizations. This project includes extending business logic & enhancing the functionalities of order management . It includes change request made by clients and doing desired changes. This project is based on IBM Web sphere Commerce Server and high availability solution built on the Java - Java EE platform using open standards, such as XML, and Web services. . It uses DB2 as its backend.Responsibility:Coding, resolving technical and design issues, unit testing, integration testing, regression testing, functional testing.