Oct 2020 – Jan 2022 · 1 yr 3 mos

Mar 2019 – Sep 2020 · 1 yr 6 mos

• ◘ Leading the cloud enablement journey for one of the largest group of ports and logistics company in Asia:
• 15+ legacy databases containing TBs of transactional data
• Seamless ingestion of GBs of Real-time and near-real-time data sent out from thousands of IoT devices
• Building a SSoT (Single Source of Truth) by consolidation of company level data from multiple source systems in cloud
• Design and development of the data lake architecture and modelling
• Going the Amazon way by implementing service-oriented architecture (SOA) at the organization level, preparing the company for the coming API economy where, exposing self-service, partner and public APIs to stay competitive in the industry will become a necessity
• All of this on Amazon Web Services
• Working closely with Solution Architects and Technology Specialists from AWS, Microsoft and Oracle and helping them with valuable ideas and feedbacks on further product and services enhancements
• ◘ Led a Proof of Concept: E-commerce Process automation and implementation of Data Lake on cloud for one of the largest FMCG clients in Asia:
• Automation of the end to end process from collecting, on a regular interval, the e-commerce metrics data generated, processing them in cloud, storing and then finally pushing them to semantic layer for reporting purposes
• Moving operational data (e.g. sales figures, transactional information, delivery records and revenue numbers) from SharePoint cloud and on premise stored manually generated csv data to AWS S3 in highly scalable columnar data formats and Redshift for generating quick insights
• The POC not only enabled the client to become cloud native but also enabled them with confidence on the process automation
• PwC won the project

Dec 2017 – Feb 2019 · 1 yr 2 mos · Bengaluru Area, India

• ◘ Built the first Autonomous AI product, to automate the grinding tasks of a decision scientist that consume much of their time. Autonomous AI was designed in a way that allows users to automate their tasks while also giving them complete control over the process through an interactive interface.
• ◘ Built large scale real-time data processing system for a leading bank with the latest Spark 2.3 and Kafka 0.10, using structured stream-stream joins on humongous amount of transaction, customer and product data and producing real time evaluation out of it to make better business decisions.
• ◘ Built EDI parser for a global technology management partner, helping huge amount of EDI being dynamically parsed following certain standards which would then be fed into an intelligent AI interface for normalization and applying rules among many other steps.
• ◘ Built end-to-end and seamless Data Science solution for the Big Brain product using scikit-learn and for Big Data using Spark ML and MLLib. It is currently being used by top banks, insurance companies and one of the largest credit card issuer in the world.
• ◘ About Razorthink:
• Razorthink is an Enterprise AI Systems company that powers organizations with Deep Learning intelligent systems that drive superior decisions, highly efficient processes and ideal outcomes. Razorthink Expert AIs include Customer Predictions, Intelligent Recommendations, Fraud Detection and Intelligent Process Automation. Founded in 2015, the company is privately held with headquarters in the San Francisco Bay Area and Research Labs in Bangalore, India.

Jan 2017 – Dec 2017 · 11 mos

• ◘ Client: North America's largest gaming retail company.
• ♦ Technology stack: Spark-SQL, Spark-Python, SnapLogic, Linux, Oracle, SQL Server.
• ♦ Cloud stack: AWS (EC2, EMR, S3, Athena, Lambda, Glacier, RDS, DynamoDB, VPC, IAM, CloudWatch, QuickSite, CloudFormation, CloudTrail, SQS).
• Around 16 different variety of data sources, from structured to unstructured, from SQL server to Google Drive.
• Using SnapLogic for data ingestion and scheduler tool to extract data from variety of data sources and putting them in the Landing area.
• Segregating the data into Hot, Warm and Cold data as per the requirement of the client, after necessary initial transformation jobs using PySpark, picking up from landing and dropping in the Raw area.
• Serious security measures taken for both Data at rest and Data in transit.
• Data compression with Streaming codecs.
• Further distinction of raw data and putting it in discovery area.
• Data Transformations (KPI Calculations) running Spark on top of AWS EMR and storing the transformed data in Refined pool buckets.
• Pick the transformed data from refined pool and convert it to the required formats for the semantic layer for smooth reporting.

Jan 2016 – Dec 2016 · 11 mos

• ◘ TPM project for a leading beverages company using Big Data bleeding edge technologies.
• Client: North America's largest beverages client (world leader in beverages).
• Technology stack: Spark-SQL, Spark-Scala, AWS EMR, HADOOP, Hortonworks Data Flow (HDF) powered by Apache NiFi, AWS S3, AWS Redshift, Hive, Sqoop, Linux, Oracle, SQL Server.
• Using NiFi or Hortonworks Data Flow as a data ingestion and scheduler tool to extract data from Oracle and SQL Servers.
• Creating the Dimension tables out of the source tables using Spark.
• Sending transactional data through downstream into master data server using NiFi with Sqoop.
• Creating the Fact tables using the dimensions and other data from SQL Server and Oracle.
• Loading the Fact results from HDFS to S3 and then finally to Redshift using NiFi.
• Applied delta logic for weekday loads.
• Did performance optimization to fit all the Facts and Dimensions execution time into the scheduled time bandwidth.
• ◘ Added Roles and Responsibilities -
• I) Project Quality Lead: December 2016 - March 2017
• Apart from the core project work, Aakash was not only the PQL for his project, but also represent all the Big Data projects as the SPOC for Quality. His roles as a PQL are -
• a) Work Product Reviews
• b) Final Inspection
• c) Tracking and Closing Audit findings
• d) Support PM/SDM or Process Analyst in Process Implementation.
• II) Innovation, Productivity & Automation Lead: January 2017 - December 2017
• Aakash was assigned with a role as the IPA Lead from Big Data practice in our Innovation Club. The basic roles and responsibilities he was assigned with are -
• a) Reusability of the innovative ideology across feasible Big Data projects,
• b) Enhancing productivity of tools and technologies used and hence reduction of manual work,
• c) Supporting automation of existing processes by tweaking the technology stack being used by respective projects,
• d) Ensuring automation process is made a highly reusable component.

Sep 2015 – Jan 2016 · 4 mos · Pune Area, India

• Got dotted into the Net.
• Capgemini acquired IGATE in the beginning of 2016.

Jun 2014 – Jul 2014 · 1 mo

• Learning how to work on live networks and then creating an End to End Enterprise Network.

Jan 2012 – Feb 2019 · 7 yrs 1 mo

• Leadership | Fundraising | Project Design and Planning | Project Implementation