Gulshan Wadhwani

I was part of team that supported all Store Systems primarily to aid in Checkout process.

• Worked on potential solutions to migrate Payment Terminals to new provider (Elavon, Verifone, Square).
• Worked on defining a roadmap to the migrate Elavon based payment Terminals to Verifone Payment Terminals (PSDK, and retail API)
• Worked extensively on building a PPL (Payment Processor Library) to use Verifone PSDK 3.64 to integrate with verifone M440 Payment Terminal to do all transactions (SALE, AUTH, REFUNDS, VOIDs)
• Created a POC to take customer inputs for Loyalty sign up process on Verifone M440 Payment Terminal, to help define a roadmap to eliminate Tablets used in the Stores.
• Evolved the Customer Input POC to capture Customer Donations for various Charitable organizations.
• Created a Listener process to query Payment Terminal for transactions done while Payment Terminal was offline. This was done to make sure business continuity even if there is n/w outage.
• Supported PWA app (Mobius) deployed on TC52 devices paired with Square Payment devices. This was developed using NodeJS, reactJS for frontend and deployed in docker containers on AWS platform (ECS)
• Worked on Minor enhancements to transition from Sales Tax to User Tax, Bundle up Price for Products, Installations, and Fee.

The Client is starting a B2B initiative, and as part of this initiative they need to create an internal tool to be able to test the normal flow of the main site. Worked with a small team of 3 resources, to deliver some major wins in 2 months, during this time I was instrumental in core development as well along with management of resources, and deliveries. The choice of Technologies here used were MERN/MEAN Stack

• Created and deployed End to End React JS(Electrode) based pages meant as alternate flow for B2B Clients. Main consumer for this was internal teams including developers, testers, BAs, Product Owners.
• Worked as lead, to capture requirements, provide design solutions and help folks to onboard and help achieve the goals.
• Extensively used Node JS with Electrode for some of client confidential services at Back end with Kafka and MongoDB for persistence and communication. These were created and deployed from scratch on Company's datacenter based infrastructure and PM2.
• Worked on micro service to check health of golden items (used for internal testing). This included integration with gTeflon APIs.
• Worked on Support role for some legacy POS Terminals using Java/J2EE with spring framework 4.x and verifone Payment Terminals. This involved helping with production issues and defects

At Pizza hut the CRM (customer relationship management) platform was migrated from Salesforce to Braze. For this implementation the project received an award from MongoDB, Snowflake and Braze for a unique Customer 360 implementation. Due to nature of integration I had chosen to implement this in MERN/MEAN Stack

• Captured and documented current state of CRM Solution used by Marketing team. This included series of meetings, identifying MVP that was needed for success.
• Capacity planning for CRM Solution based on all customers (approx. 368 million customers) vs 52 million (29 million loyalty and 23 million non-loyalty customers). This included about (42 profile attributes, and 40 CAF attributes). Capacity planning was done across multiple areas. Including MongoDB, Snowflake, AWS Services that’ll be used, and credits that’ll be consumed in Braze.
• Designed and developed REST APIs(Node.js, Express.js with MongoDB) for providing coupons/deals/filtered deals for Braze. This was designed to handle capacity of 250k calls/min. This solution was tested using Redline - JMeter scripts. Test data was created using ad-hoc code in Node.JS to generate records based on schema. Business driver was to target 20 million + customers within 3-hours lunch time window, and dinner time window including all 4 time zones.
• Created end to end solution for curbside pickup at Pizza hut, this included capturing order status from SUS/NPC and sending SMS to customer and when customer responds with one of approved messages inform store. Also dealt with appropriate negative scenarios and audit trail for legal purposes. Integration was done using AWS Pinpoint (SDK), Twilio (SDK) and Braze ( REST API). All 3 SMS providers were included as this SMS-handler service was further enhanced to support any corporate needs to engage customers using SMS.
• Designed and developed REST APIs for customer profile/CRM-IDs, for capturing customer communication preference, customer profile. This also included design of MongoDB Collections. This API was developed to handle up to 5k calls/min. Over period time it was adjusted to handle 11k calls/min during Friday peek loads in April-2020. This also published updates/changes to Kafka for any downstream consumers.
• Designed and developed Process to Sync Computed attributes from Snowflake to Mongo 4.0 for approximately ½ million customers each day. This was also planned to sync 50 million customer sync in under 12 hours. This posed significant challenge and extensive solution to make sure real time APIs are not impacted.
• Worked on new project initiatives, including streamlining purchase events, to CRM platform to enable marketing team to tweak their campaigns and benefit from orders made my customers.
• Worked in a team setting of 4 Developers, and 2 QAs starting with onboarding, steer and deliver the project in given aggressive timelines.
• Helped implement text-to-join feature with team, so that we identify and correlate customers and target them. This was implemented using lambda layers.
• Redesigned few services to improve performance, this was done using ElastiCache layer (Redis), we cached essential data about Products, coupons, stores, strategies to better capture essentials and send data to CRM platform.
• Designed and Developed Kafka Message Consumer using Node.JS, to capture updates to profiles and update braze accordingly.
• Designed and enhanced Kafka Consumer, to capture Phone Numbers, and key fields to set Alias(es) to Braze profiles, and set them to SMS Subscription group
• Created Lambda functions, to capture Un-Subscribe email/SMS events coming from Currents, and update profiles in MongoDB.
• Created AWS Lambda functions in Java 8 to handle requests for Loyalty Customers, wherein SUS/NPC or call center made calls to find customers based on phone number, or email address.
• Created AWS Lambda functions in Java 8 to handle requests from external systems, to provide CRMIDs based on punch sys id (from Punchh System), qo_cust_guid (from Quick Order System), This was quick and dirty solution to aid in quick assessment of integration with these systems. Plan was to enhance them into full-fledged microservices.
• Created Batch Jobs to fetch/load provided files from AWS S3 for “reject-keywords”, and “approved-names” and tag each profile suitably if first name is suitable or not.
• Created Shell Script based process to monitor and alert if Messages from Kaka topic lag by more than designated threshold
• Created Shell Script based process to monitor and Alert if Jobs are delayed by more than 10 minutes.
• Created Process to capture snapshot from Braze and Sync Mongo and Snowflake. This was done to aid in automated testing.
• Created Process to upload Coupons (from QO) to MongoDB. This involved loading 20 million coupon codes in less than 1 minute
• Designed and developed process to offload used coupons from MongoDB to S3 Location
• Designed and developed process to Capture NFL Schedule from official website and send relevant campaigns to fans.
• Implemented end to end solution for CI/CD using Circle CI (Mono-repositories), and Terraform 0.11 to deploy artifacts (ECS Services, S3 Buckets and IAM profiles). This was further migrated to Bitbucket pipelines
• Migrated the MongoDB Cluster from 4.0 to 4.20 to 5.0 version.

• 7-Eleven is exploring solutions to replace their store specific ISP (in-store processors) and POS devices so that everything can be hosted on cloud
• As AWS Cloud SME, captured critical path for current application, meaning most read operations and write operations in application
• Built one application using MEAN Stack (Node JS, Express.js) mimicking same flow (read and writes), and for POC (only items, inventory, departments and Item restriction tables) were migrated to Dynamo DB, MongoDB, and AWS RDS(PostgreSQL)
• This was done to aid in estimating cost to implement, cost to maintain (from operation standpoint) and performance (from feasibility standpoint)
• Finally, this application was load tested using JMeter based scripts on EC2 instance to compare results.

• At ULTA Beauty, built new applications for store services, this includes salon services, events and promotions
• Worked on New features/stories allocated during sprint cycles this included developing backend APIs using MEAN/MERN Stack) right from point when customer enters store
• Worked with team to refine data model and do stress testing across multiple functional areas
• With results from Stress testing defined indexing strategy for tables
• As single point of contact, implemented High availability/Failover PostgreSQL Database cluster from scratch (v9.5)
• This was initially done using pgpool-II in QA and Stress servers (with streaming replication)
• Later this was changed to EDB Failover Manager (streaming replication) as client procured licenses
• Implemented routine jobs (for Backup and recovery, managing logs and WAL files) in cron, this is keeping PITR in mind
• Implemented Monitoring tools including PEM (PostgreSQL enterprise manager) for 2 instances (production and non-production)
• Installed and configured postGIS for sprucebot application
• Installed and configured redis DB for client

• At AIM Specialty Health, they have couple of applications to analyze and automatically approve/disapprove requests for tests based on certain algorithms. I worked as a Lead developer and DBA for their 302 project. Herein we defined the collections (in MongoDB 3.4) and tables (in PostgreSQL 9.5) that were central to RAD-Oncology, completed the capacity planning and cost estimation This included setting up the infrastructure(on-premise) for application and create POCs and finally deploy them.

• At VCE they build virtually converged platforms and deliver to clients ready to use vBlocks, vxRails etc To capture requirements, configure, quote and price these ready to use platforms, they have couple of web applications utilizing, SQL Server 20012, PostgreSQL 9.4, MongoDB at backend, for front end they are using Java 8, Angular JS, Node JS and Express JS (MEAN Stack)
• I was also instrumental in building and supporting REST APIs developed using MEAN Stack (Node JS/Express.JS), J2EE 1.7, and Spring framework to provide the current customers/vendors for the VCE tools to configure their servers, this was also being developed to support them so that customers can do a similar configuration of their VBlocks like we can configure our laptops, but this not only configured the hardware but also configure the networking options they had selected. The final product created ready to use servers that can be used for application deployments.

• At SeaWorld they have been using a Bill Inmon styled Corporate Information Factory, with some DataMart’s with very limited information about the atomic sales data, and majorly Aggregates. They intended to implement a complete corporate Data warehouse based on Ralph Kimball methodology.
• We had started a project to onboard integration with 3rd party to be able to sell packages event packages, vacation packages. This Integration involved supporting APIs based on the datawarehouse based metrics for recommendations. I created the POC for this using Node.JS/Express.JS and SQL Server at backend.
• I had also created Subscribers for a webhook exposed using MEAN Stack (Node.JS/Express.JS and MognoDB). This was to capture events generated when packages were added to cart, and purchased. This was to further add to the Datawarehouse to asses the trends and likability of the packages.

• At BCBSM they have implemented CPDM data mart, that houses data ranging from basic subject areas like members, Providers, to data about Claims, Healthcare indicators etc totaling to 10 TB+. This data is used for many initiatives taken at BCBSM to reduce cost of operation, evaluate and reward the healthcare providers who help them keep costs down

• At JLL they have close to 200 end-client’s contracts, and each Client specific data/Transformation is managed in their specific DB instance. They are acquiring more clients, and standardizing the way Jobs/reports, Created Transformation packages for each Client. The collective size of all DBs was at the time more than 10 TBs, and many tables having billion+ records. Worked on all aspects of data in this along with Tracker application that provided an insight into the underlying data. This Solution was built using SSIS Packages, C# ASP.NET, jQuery, Telerik controls, RadScript Manager, Rad Controls, HTML, CSS,

• The Client is having couple of applications built on Vignette Content Management, Vignette Application portal, and DSM module that use oracle 10g. Worked extensively on the application and backend (Database) to aid in split of the web platform for 2 separate companies Motorola Mobility devices, and Motorola Business solutions. This was done using Vignette 7.5 and then we migrated to Vignette 8.0 suite (for VCM and VAP), Oracle PL/SQL, JSR 168 portlets, Java POJOs.

• Client is Extracting data from GOLD (retail management system) and pushing the data into Point of Sales (POS) and Lawson (accounting systems). One part was to write jobs in Java to extract data in XML format from multiple end points including warehouse, stores, Vendor portal and place them in a shared location, there was an analytical part, wherein data was pushed into a data warehouse, to develop reports/trend lines and make suitable business decisions. Worked extensively on both ends to make sure of data consistency and avoid any staleness. This was done using AIX 4.2, J2EE1.5, Spring, Hibernate, oracle PL/SQL, SSIS 2008 Packages, SQL Server 2008

• Client is migrating their website www.mibank.com to Vignette products, SQL Server 2005 and Oracle 11i. Worked extensively using Vignette APIs, Java and SQL Server to aid in the migration, performance testing and validation of data across all pages.

• This application and data involved was a result of HRMS system to maintain records, timesheets for various consultants. It is also used by accounts to generate invoices, track finances in combination with quick books. This was done using AJAX, HTML, CSS, and ASP.

• Worked on the integration and application built in C# and ASP.NET. and the reports built using SSRS
• Created and Supported Stored procedures, views to handle synchronization process between ECARS 1.0 and 2.0. in addition to it Created and Supported SSIS Packages that pulled the data from Oracle 10g System, AS400/IBM DB2, flat files on FTP locations, and XML files. This data was transformed and loaded it into a ODS in MS SQL Server 2005.

• In this project there were couple of websites built on Vignette content management Servers, having databases in MS SQL Server 2005, and Oracle 9i. As a lead engineer I was tasked to enhance and maintain the applications. The websites included iVillage.com, iVillage.co.uk

• In this project the data from various websites, and offline sales team was extracted, transformed and loaded in SQL Server 2005 based database. This was an OLAP having multi-dimensional model having a snowflake schema.

• Project – www.otis.com - This site was built using Vignette content management server 5.6, and Oracle 8i. This included creating DB Objects, Created PL/SQL block to generate triggers on all the Vignette-System tables.
• Data warehouse/Reports
• Project – Eclipse - This was a Component management system, integrated with a PLM Solution. It is mainly used by engineering people who design elevators, Escalators, etc.