Mar 2017 – Present · 9 yrs

• As a part of the fintech team, i have designed and implemented a recommendation engine for vendor. As a part of this , I was the lead engineer building out initial infrastructure on AWS EMR, AWS Elastic search and then Hadoop using spark EMR. In this project , we consumed the S3 file database and then built the infrastructure to process these s3 file consisting of purchase order, invoices ,shipment data. We applied certain set of rules on top of them to to get the defect root cause .This root cause then surfaced on analytical application using AWS elastic search as back .
• As a part of this project , i did POC for 1) service oriented architecture 2) Data oriented architecture.
• Then finalized the the final data oriented approach .Then finally i designed to have a 3 job architecture.
• 1) PrepareResearchDataJob
• 2) ExecuteRulesDataJob
• 3) WriteDefetctToElasticsearchJob
• The system is in prod for last 2 year and 3 month.

Jul 2014 – Present · 11 yrs 8 mos

• Here I am basically working with banking module for vendor ships us the goods , they invoice us for the goods that they shipped. Now In order for Amazon to make automated payment, Amazon asks vendor to provide its banking details. These banking details are highly confidential and is being stored in terms of tokens in the amazon system. I built a workflow that would secure this process via multi factor authentication.
• This Authentication involve OTP generation, then customer id based authentication, then current account token verification and then finally rule based verification . These all steps ensured even if one of the actor i.e the email get compromised still the banking details are safe.
• As a part of this i also did 10 country launch for our banking application. especially the china banking module were a challenge as they are having completely different data policy. One such policy is to store the china related data into china DB.
• Thanks
• Kishor