Yogesh Agrawal

Yogesh Agrawal

Graduate Student | Software Developer

Contact Me

About Me

A computer science enthusiast with 3+ years of industrial work experience and 5+ years of academic experience. Currently pursuing MS in Computer Science at Stony Brook University.

I am very passionate about technology and how it is at the centre of current transformation the world is observing. I have worked on large-scale projects, whose demands spanned to breadth of scalability, availability and reliability.

Latest Projects

SfM Pipeline

Goal is to reconstruct 3D scene correctly from multiple images of a scene, in which many structures are repeated/similar. Used techniques in Visual Odometry to determine correct camera position path.

View on GitHub

Byzantine Chain Replication using Shuttle Protocol

Implemnted byzantine fault tolerant chain replication for servers. Created a working prototype using the Shuttle protocol as given by Robbert van Renesse. Used distAlgo for the creation of async clients and message passing.

View on GitHub

More on GitHub

Work Experience

Software Developer - OYO Rooms ( Feb 2016 - July 2017 )

Worked as a Full Stack Developer, implementing Front-end & back-end functionalities and features across many products.

Designed & developed internal web portal with focus on scalability. Included creation of bookings by customer service center, managing hotels database and relevant metrics, deals for customers, user/employee permissions.

Developed hybrid mobile application for OYO Owners, capable of running on Android and iOS deviceswith the same code-base.

Associate Engineer - Qualcomm ( July 2014 - Feb 2016 )

Worker as a Framework Developer and improved reliability of existing automated testing framework, CIT, from 70% to 95% by updating outdated code base and use of best practices. Currently being used for testing of Snapdragon, MDM, Wearable & IoT device chipsets.

Extended capability of framework, using concepts of distributive and parallel computing, to enable testing of multiple devices on same base station, called ParallelCIT. Reduced cost by approx $1000 per device. Planned and executed roll-out in India, China and US; including knowledge sessions on new framework to sibling teams in these locations.

Automated setup and configuration of each device on base station, reducing time to 5 mins from 1 hr.

Summer Intern - Qualcomm ( Summer 2013 )

Built online tool, Build Insight, which lists modified files across two builds of OSs with approx 10^5 files/build.