Sidharth Agarwal Email & Phone Number

Seattle, Washington, United States

Seattle, Washington, United States

Working on Uber Eats Order Platform, to enable frustration free item selection, checkout, and post order user experience. Re-defining Uber-perience for our customer - A customer experience that enables delivery of goods and service to their door step with minimal user touch points and intervention.

Sunnyvale, California, United States

Working to manage and build the next generation search engine capabilities.

Sunnyvale, California, United States

Data Extraction Pipeline Platformo Created a serverless data lake pipeline that would allow users to easily onboard production systems to extract data based on config file information to be used in applicationso Developed a platform for easy onboarding of machine learning teams/services to extract data using configuration inputo Used Glue ETL framework to create multiple jobs to publish data based on user inputo Used spark 2.0 PySpark scripts to achieve parallel computation and faster run timeo Used Step function to Orchestrate calling of jobs based on events and running batches at times to improve pipeline efficiencyo Used Lambda function to spin up several Glue job threads to run multiple user requests at a timeo The Pipeline could scale to support 300 requests per second made to retrieve and utilize data by production systemso Helped reduce onboarding overhead by 30% thus saving 4 days’ worth of SDE effort on a bi-weekly basis

Greater Seattle Area

Ticket Automation Web Service – o Created a serverless web service to help resolve tickets automaticallyo Used event driven and multithreaded programming to automate ticket resolutiono Created a generic template driven service which helped easy onboarding for multiple teams within short amount of timeo Used Rest APIs and AWS technologies that could help in scaling of the service to be able to handle 250 TPSo Deployed Service on a fleet of machines thus supporting multiple teams across Amazon reducing manual hours by 32%, thus saving ~700 hrs. on an annual basis.Costs Automation - o Increased reports accuracy by 20% and reduces manual work by 384 hrs./annuallyo Backend Java application reads data from Amazon Data Warehouse solution using SQS and SNSo Lambda Function to move data from s3 to Redshift based on user activity.o Used advanced tree data structures and algorithms to manipulate and clean dimension datao Provided flat level data to be easily understood by user for changing cost calculation business logico Validates data loaded by user reading csv datao Provides information to user via SES email service, using Amazon Internal service informing the user about success or failure while uploading the data and then send guidance to the user to resolve any data quality issues.

Seattle, Washington, United States

Downward Stream Impact System – o The new system saves approximately 480 manual hrs. on an annual basiso Backend Python application reads data and transfers it between s3 bucketso Decrypted data on move and encrypted data at rest in order to load data into redshift.o The system spins up an EMR cluster to load 2PBs of data on a daily basis to run millions of calculations to be used in machine learning models.o Used CloudWatch alarms and internal Amazon ticketing system to make the System operational ready. Net Present Value Automation - o Saves 3 hrs. daily and reduced time to completion for customers from 2 days to a couple of minuteso Backend Java application reads issues from ticket using SNS and SQSo Reads and filter SQS messages in Applicationo Processes messages and provides interactive suggestions to customer to solve any issues in the ticketP&L Application -o saves ~784 manual hrs. on an annual basiso Java Application using SQL database redshift in the backend to store and create reports on the fly.o Designed two applications - optimized for Space and time complexity, due to lack of hardware, the team utilizes these desktop applications based on their hardware capability.o Different format reports can be built using java application, user can choose different inputs to customize their reports such as year, scenario, type and other. The application spans across multiple teams within Amazon.o The current system design is capable of handling 50 TPS (based on customer requirement)

Phoenix, Arizona Area

a) Enhanced business decision making capability by handling data quality issues on 4 million recordsb) Increased SQL Query efficiency by reducing execution time by a factor of 10c) Identified trends & Patterns by carrying out descriptive analytics on student's demographics featuresd) Performed Stepwise variable selection method to select 12 factors out of 200 factors for building "Graduation Model"e) Created and examined executive dashboards,scorecards and glossary to identify gaps for Human Resourcesf) Created Workflow for pushing data from People Soft Materialized View to Data Warehouse Tables which in turn would help to save lives and economy destruction at the time of an emergency

Tucson, Arizona Area

a)Proposed Data Mart design conforming to the Enterprise Data Warehouse standardsb)Mapped business metadata to technical metadata in order to convert user requirements into dashboards and reportsc)Converted data into information and actionable insights to identify new opportunities for branding Qualcomm Imaged)Transformed product transactional data into aggregate sales level data across partners and regions for easier analysise)Provided recommendations and suggestions in order to redefine Strategic alliances with various OEM partners

Tucson, Arizona Area

a) Works and coordinates team of 8 UG-TA's and 47 preceptors to manage the Lab sessions for the MIS111 class consisting of 1800 studentsb) Assists student in basic financial analysis (stock portfolio analysis), business management(franchise operations) and marketing (website development)

Client Novartis | Patch Academy|| User Experience | Data Analyst• Participated in Requirement Gathering, Analysis, Design, Coding and Configuration Management. Worked on SQL and Objective C in order to create a responsive user interface as per client needs and requirement.• Enhanced User experience by analyzing user application interaction data such as bounce rate, click streams to help stakeholders evolve their product's design and features

Software Solution Client |Mobile Application (Internship Project at Infosys)|Android Project | Lead• Optimized utilization of resources; quality control by monitoring and reporting progress in a timely manner. Created an Android application for logging and reporting of service requests by user.This project was nominated as the best intern project.

Management Information System, Business Intelligence

Bachelor'S Degree, Computer Science, A

Activities and Societies: Debate Team, JYC, Robotics* Appointed as External Affairs Minister for the Jaypee Youth Club. * Organised.