Mohit Garg

New York, NY · mogarg@protonmail.com

Hi, I’m Mohit, a Software Engineer at Bloomberg LP. I have worked on improving Distributed Systems, led Mobile Development, and I have a strong interest in building scalable solutions to solve complex problems.

I am currently working as a Software Engineer at Bloomberg LP. As a Masters in Computer Engineering at Virginia Tech, I worked on researching protocols that power Distributed Systems like Cloud and Blockchains for high throughput, global scalability, five 9’s of availability in the presence of crashes and malicious behavior. My thesis introduced a new Byzantine Fault-Tolerant Geo-Replication Protocol with numbers that achieve 3.5x over industry standards.

Before joining Virginia Tech, I worked for three years at Indus Valley Partners which provides award-winning software solutions for the Hedge Fund Industry. My love for Mobile First motivated me to help set up an iOS development team. We created Nebula a secure, highly customisable iPad application which enables Hedge Fund managers to study their analytics on the move. Working on this platform made for a rich understanding of framework conventions, best practices, memory management, and development tools. I also learned a lot about Finance and fell in love with Cryptography as a means to secure highly sensitive data.

I have since found passion and has worked on distributed data processing and search technologies such as Kafka, Hadoop,Elasticsearch.

Check out some of my Projects for more. :)

Skills

Architecture

IP Networking
DNS
Firewalls
Load Balancing
Microservices
RESTful APIs
SaaS/PaaS/IaaS

Languages and Tools

C/C++
Python
Go
Java
TLA+
MySQL
Git
Solidity
Bash/Zsh
Dtrace/BPFTrace
Splunk

Platform

iOS
Linux
AWS
Elastic Search Stack
Hyperledger
Ethereum

Databases

MySQL
MongoDB
PostgreSQL
Apache Lucene

Containers & Cloud

Projects

Elpis

A Multi-leader Cross Fault-Tolerance Algorithm

Apollo

Apollo Ticketing Management (Smart Contract for Hyperledger Fabric Blockchain)

Wait! There's more..

See all Projects for more examples!

Experience

SOFTWARE ENGINEER

Bloomberg LP

January 2019 - Present

Graduate Research Assistant

Systems Software Research Group, Virginia Tech

Researched core protocols that power Distributed Systems like Cloud and Blockchains.

Improved on current industry standards to achieve high throughput, global scalability, 99.999% availability while tolerating crashes and malicious behavior.

Designed Elpis, a multi-leader Byzantine Fault-Tolerant Consensus Protocol (submitted, pending review) which outperforms competitors by 3.5x and cuts latency by 50% in Geo-Replication.

Improved availability under high client loads by solving the livelock problem in M Paxos by designing a Collision Recovery Protocol which enables operation with 100% conflicting commands as compared to 10% earlier.

Implemented a Distributed Key-Value Store based on MPaxos, PBFT, and Zyzzyva in Java which provides throughputs to the order of 10 ops/sec.

January 2017 - August 2018 (1.5 years)

STUDENT DATA ENGINEER

IT Security Office, Virginia Tech

Indexed 1 million rows of daily network log data by implementing a Kafka based pipeline for storage in the university's Elastic Search Stack

Discovered usage patterns using Spatio-temporal modeling of log data for network planning and security.

May, 2016 — December, 2016 (7 months)

SOFTWARE DEVELOPER

VIRGINIA TECH

Developed an iOS application in Swift with a data dashboard which provides custom forms, plotting, analytics, and sharing capabilities

Enabled 15+ employees to collect sports data in the field and generate reports.

March, 2016 — August, 2016 (5 months)

SOFTWARE ENGINEER

Indus Valley Partners

Founded the Mobile Development Team along with a fellow engineer by landing deals with 3 major Hedge Fund clients and developed multiple enterprise iOS apps for them

Optimized the device power consumption for real-time data by devising query processing and caching strategies, extending battery life by 20

Secured sensitive data on all 40+ operational devices by developing a framework to utilize Common Crypto for on-device data encryption

Implemented high-performance arithmetic calculations by utilizing SIMD instructions, allowing apps to achieve 4000x speedup for 1 million entries.

June, 2013 — October, 2015 (2 years 4 months)

Education

Virginia Tech

Master of Science

Computer Engineering

2016 - 2018

Punjab Engineering College

Bachelor of Engineering

Electronics and Electrical Communication

2009 - 2013

Blogs

Rapid: Distributed Membership Service at Scale

This article presents a summary of the paper by Lalith Suresh, Dahlia Malkhi, Parikshit Gopalan, Ivan Porto Carreiro, and Zeeshan Lokhandwala, which appeared in USENIX Annual Technical Conference 2018. I also presented this paper in the Distributed Systems Reading Group, a weekly public paper reading group hosted by Prof Murat Debirmas. This paper primarily about membership in distributed systems in the presence of gray failures. While crash faults are relatively easy to detect and fix, gray failures often go undetected or take a long time to localize and fix.

May 24, 2020