Sandip Mahanta

Created a team of engineers that builds the home grown last mile delivery system from scratch for Walmart Online Grocery's home delivery. I was responsible from hiring the first set of engineers to launching it and scaling to 25 million deliveries per year.

• Build the Driver management system; ecosystem for crowd sourced driver. Starting from onboarding drivers, maintaining records, triggering payouts to driver and maintaining availability of drivers for deliveries.
• Build live order tracking system for customer tracking purpose
• Build the driver suggestion system that gives the best fit driver for a give delivery.
• Build the driver rating system that rates each driver based on their performance over various parameter such as on time delivery and on time pickup.
• Build the driver matching platform that lets the business to configure various model to search a driver given time duration. It was further enhanced to pick right kind of driver matching template based on time to deliver.
• Scaled the system to deliver grocery for Walmart to 60% of US household
• Focused on reducing cost per delivery by implementing batching of multiple orders in single trip.
• Due to Covid adapted the system and team to new norm and launched multiple features in quick time like contact less delivery and extended the system to delivery GM deliveries for Walmart
• Expanded the system to multiple markets - Mexico, Canada and UK
• Overall technologies used - Java, Oracle, Maria DB, Elastic Search, Cassandra, Apache Spark

I build the home grown sourcing system from scratch that provides a service to ‘Precise Delivery date’ to the e-commerce customers on Walmart.com site

• I lead a team from tech perspective to deliver on sourcing supply chain projects. I have successfully lead the team to deliver free 2-day shipping for Walmart US. I have also delivery multiple critical projects such as “slow ship discount”, “pickup discounts”, “Geo based sourcing”. I have also lead the product to successfully scale for multiple peak holiday season sales.
• I designed and made the sourcing engine of Walmart as a generic product that can be consumed by any other site of Walmart or any other country. Ongoing projects are targeted to reuse the same for Sams Club, Jet.com and Walmart Canada.
• Worked on multiple performance enhancements such as Cassandra tuning, application profiling, reducing garbage collection pause time, etc to make the product scalable.
• Overall Technologies worked on - Java, Cassandra, Apache Spark

I lead the sourcing team that is responsible for promising best delivery date that we can keep.

• Sourcing engine [USA]: Build the home grown sourcing engine for Walmart from scratch that replaced the legacy-sourcing engine. The new build is faster, scalable and adaptable to Walmart business scenario.
• Integrated the application with Kafka – elastic search – Kibana ecosystem for analytics required for business decision and operations

Sourcing engine [International]: I Lead a team of 4 in technical front that aims to launch our product in United Kingdom. I process requirements with product managers at UK; plan, design and implement any change that is required in the product.

• Along with new functionalities, I re-designed the product to support multi-tenant i.e. Same product / code / deployment can be used to serve multiple countries

Technologies worked on - Java, Scala, Hive, MapReduce

• .As part of sourcing system I worked on Packaging Algorithm. Implemented a Best Fit Greedy algorithm that solves 3D packaging problem for Walmart warehouses and helps in calculating precise delivery date. The algorithm optimizes the number of boxes used and reduces wastage of space.
• Implemented pruning of combinatorial algorithm to generate all possible solutions for sourcing. This algorithm analyzes all items to distributor combination that can be sourced, and optimizes on various parameters such as consolidation, cost to company and earliest date in order to provide the best solution.
• Technologies worked on - Java, Cassandra

Retail Business Services: Integrate ticketing operational workflow with different external system e.g. pricing, purchase order, catalog, etc. reducing context switches for Amazon catalog associates leading to increase in productivity.

• Designed the solution as Data Access Object pattern which provided abstract interfaces for all external system
• Implemented the entire project of integrating external systems with ticketing system of Amazon.

Vendor Central (Vendor Item Setup): Auto approval of New Item Setup Problem Statement: Removed all manual touch points in New Item Setup process of Amazon

• Automated all kinds of manual touch points done by Amazon associates by integrating with Amazon’s multi lingual dictionary and data error classification and automation.

Retail Business Services: Built a reporting framework for Amazon Retail Business Services to serve the entire reporting requirement needed to make effective business designs for Amazon Retail.

• Built more than 30 different kinds of report across different business with Amazon providing data through email, S3 downloads, windows share. Tools supported xls and csv file format.
• Technologies: Java, Tomcat, POI, JSP, MySQL, Hibernate

Retail Business Services: Enable all backend data provision to enable Oracle Business Intelligence tool.

• Integrated various data sources in amazon with Retail Oracle database with proper frequency and reliability enabling building Oracle Business Intelligence tool on the same database

RBS Timesheet: Build a New Item Timesheet webpage for Retail Business Service of Amazon to get progress metrics for every user across all Countries.

Amazon Flexible Payment System: Migrated the code base of Flexible Payment System from Mason/Subversion to Amazon build system and Perforce.