Rohit Mishra
Software Engineer
Education
University of Texas at Arlington
August 2017 - August 2019
Master's Computer Science
Coursework: Design and Analysis of Algorithms, Machine Learning, Parallel Processing, Robotics, Convex Optimization, Internet of Things, Design and Construction of Compilers
Major: Master's Thesis in Systems Architecture | GPA 3.87
Skills
Languages: Python, Java, JavaScript, TypeScript, Bash Scripting, Powershell, C/C++
Tools and Frameworks: Spring, Angular 6, React, Django, Cassandra, Dialogflow, Git, AWS, Serverless
Professional Experience
Amazon
Seattle, WA
Software Development Engineer
September 2019 - Present

Responsibilities

  • Translating business and functional requirements into concrete deliverables with the design, development, testing, and deployment of highly scalable distributed services on AWS.
  • Designing, developing and maintaining core system features, services and engines.

  • Help define product features, drive the system architecture, and spearhead the best practices that enable a quality product.
  • Operating in an Agile/Scrum environment to deliver high quality software against aggressive schedules.

Fiserv
Dallas, TX
Software Developer Intern
June 2018 - August 2018

Responsibilities

  • Writing application performance summary and code documentation for use by senior engineering staff to determine defects in the application. 
  • Developing a Java based command-line tool using Test Driven Development (TDD) to automate major daily tasks.

Achievements

  • Prototyped a distributed archival project using Apache Cassandra for storing and retrieving any kind of file with response times of <1 second.
  • Solved a crucial business problem by exposing the entire project using a micro-services based REST api for reuse across the organization.
  • Used Test Driven Development (TDD) to produce high-quality code and correctly estimate milestone timelines.
  • Uncovered issues in production web services using JMeter load testing and JMC performance profiling.

Tech Mahindra (AT&T)
Pune, MH, India
Software Engineer/Full Stack Web Developer
August 2015 - July 2017
  • Migration of a critical application with monolithic architecture (10,000+ users) to newer micro-services based technology stack for improved responsiveness and scalability. Also presented performance profiling reports of before and after improvements using industry relevant metrics.
  • Mitigation of issues in production environment while minimizing impact to customer.

  • Solved multiple issues within SLA for critical applications with 10,000+ users.

Projects
Multiprocess producer consumer Queue (Linux/C/C++)
May 2018 - Present

Masters Thesis project on a framework for a multicore approach to solving the generic multi-producer multi-consumer problem.

  • Uses cooperative lock-less queuing/dequing between threads to reduce contention.
  • Operates in shared memory. Hence can be used for Inter-Process Communication.
  • Lowered tail latencies of existing lock based system from <10 us to <6 us.
  • Improved current lock-free system which reduced tail latencies from <3 us to <800 ns.

Histolomaps (Javascript) U-HACK MED 18
November 2018 - December 2018

An efficient annotation storage/visualization/editing framework to visualize deep-zoom tissue images. 

  • Developed using the popular 3D Javascript engine Three.js which can automatically tap into onboard GPUs using WebGL.

  • Support for annotating/editing upto a 1,000,000 objects on the tissue image. 
  • Main aim is for machine learning algorithms to be trained to annotate cells (e.g. cancer cells) via reinforcement learning.

Music Speech Classifier (Python)
March 2018 - May 2018

Classifier for distinguishing between audio samples of Speech and Music.

  • Dataset compiled from a combination of the GTZAN Music/Speech collection and Columbia Music/Speech dataset.
  • Tested the accuracy of a variety classifiers including Decision Trees, Naive Bayes, KNNs, SVMs and Neural Networks.

  • Final Ensemble classifier accuracy 96.53 %
Word Assistant (Python)
October 2018 - November 2018

Google assistant plugin for Microsoft Word.

  • Completely control all aspects of a Word document like creation, editing, deletion, etc.
  • Can easily be extended to control other applications in any desktop environment (Linux/Mac/Windows).
  • Deployed using Dialogflow and Heroku. Hence can be triggered on any Google Assistant device by saying "Talk to Microsoft Word".

Activities
Center for Advanced Computing Systems (CACS), UTA · Graduate Research Assistant
November 2018 - Present

Performing research on topics related to Systems/Architecture.

Department of Computer Science · Graduate Teaching Assistant
August 2018 - Present

Data Analysis and Modeling Techniques/Intro to C Programming