Sourav Ravikumar

Computer Science at UC San Diego · johar.sourav97@gmail.com

I'm a master's student at UCSD who is highly motivated to innovate in the field of AI. I aim to build artificial agents that show human-like “common sense” in their operating domains. When I'm not in front of my computer, I like to hike, play the piano, and read a book. I also upload programming tutorials on my YouTube channel where I wish I was more active.

Education

Univeristy of California San Diego

Master of Science
Computer Science

GPA: -/4.0

Sep 2021 - Jun 2023 (expected)

Amrita Vishwa Vidyapeetham

Bachelor of Technology
Computer Science and Engineering

GPA: 9.46/10

Aug 2015 - Jun 2019

Professional Experience

Member of Technical Staff 2

VMware R&D

I worked for AppDefense (now CarbonBlack) which is an application security product for workloads and endpoints. Part of my work involves designing and building automation frameworks in Python to ensure correctness of our security features. I also built a Vulnerability Assessment for determining the vulnerabilities present in OSes and applications which saved the company a ton of money because we were earlier using a third party service for the same.

Jul 2019 - Aug 2021

Intern

VMware R&D, NSX

I was responsible for re-writing the lower level code for an internal framework to comply with newer standards. As the framework was in constant use by over 50+ engineers, I had to make changes such that they do not impact anyone at the user-level. My role required strong Python skills, shell scripting and working over virtual machines and datacenters.

Jan 2019 - Jun 2019

Digital Technology Intern

GE Digital

As a technology intern under the GE Renewables umbrella, I optimized the algorithm used for determining the mechanical load on GE Wind Turbines. I worked along-side principal scientists and brought down the run-time from 3 hours to 20 minutes by using efficient mathematical minimization functions.

May 2018 - Jul 2018

Research Experience

Undergraduate Research Assistant

Amrita Vishwa Vidyapeetham

Alongside my corporate intersnhip with VMware, I also participated in an academic research project with Dr. Vidhya Balasubramanian and Dr. Shunmuga Velayutham. We outlined the design of a successful fitness function, which "evolves" an artificial agent to play (and perfect) the Snake Game - using a novel combination of Neuroevolution and Search Algorithms (like Curiosity Search and Novelty Search). We then modeled the snake game as a goal-seeking problem where our agents showed great adaptability even in uncertain environments.

Jan 2019 - Jul 2020

Research Intern

Preflet

I served as a part-time research intern at Preflet, a startup in Lisbon, Portugal. I built a platform which automatically trains a Machine Learning model using the most suitable algorithm given a dataset. By using population-based Evolutionary Algorithms, the platform was able to fine-tune the model's hyper parameters which further increased its robustness.

Feb 2019 - Nov 2019

Undergraduate Research Assistant

Amrita Vishwa Vidyapeetham

Under the mentorship of Dr. Gowtham R and Prof. Dayanand Vinod, I designed a digital and real-time solution to prevent Human-Elephant conflicts across the forest fringes of Kerala, India. We developed a novel layered approach that minimizes the load of running deep learning models on computationally limited devices like Raspberry Pi. The system was successfully able to detecte all 4 instances of intruding elephant herds along our campus boundary.

Oct 2018 - Jan 2019

Publications

  • Vignesh Kumar K, Sourav Ravikumar, Shunmuga Velayutham C, Vidhya Balasubramanian
    Fitness Function Design for Neuroevolution in Goal-finding Game Environments,
    Springer’s Communications in Computer and Information Science, 2020, DOI: 10.1007/978-3-030-63119-2_41

  • Sourav Ravikumar, Dayanand Vinod, Gowtham Ramesh, Sini Raj Pulari, Senthilkumar Mathi
    A Layered Approach to Detect Elephants in Live Surveillance Video Streams using Convolutional Neural Networks,
    Journal of Intelligent & Fuzzy Systems, 2019, DOI: 10.3233/JIFS-179710

Projects

Snake AI

Go Get Tested

HEC Avoidance System

Handy

Auto Attendance

Other Projects

Awards & Honors

  • Honoured Member of the Tech{know}logy Club, Amrita.
    Conducted two workshops (Chatbots 101 and Hands-on Machine Learning) which were attended by more than 250 people totally

  • Received the Best Mentor Award at VMware’s Bootcamp Hackathon, 2020 - mentored the winning team of 4 new-grad hires

  • Youngest author to publish online courses for professionals and students at Packt Publishing, UK

  • Finalist in the Amrita Annual Public Speaking Contast, 2018

  • Won the 1st place in the CTF Event organised as part of the MooseCon Web Security Conference, Pune, 2019.