About Me

I am a Mechatronics bachelor with a great passion towards Artificial Intelligence.
My project and work experiences mostly fall under deep learning, computer vision
and natural language.

Besides work, I participate in Kaggle when I get free time. Cooking and spending
time with friends have been my stress reliever.

  • Concepts
    Deep Learning
    Computer Vision
    Natural Language Processing
    Data Structures & Algorithms
    Digital Signal Processing
  • Coding Languages
    Python
    C++
    Matlab
    HTML
    CSS
    Git
  • 2016 - 2018
    Worked on Computer Vision at SRM Scro/Drona
  • 2019 - 2020
    Computer Vision Intern at Sigillieum Software Consultants
  • 2021 - 2022
    Machine Learning Intern at Slimmer Ai
  • 2015
    Passed high school at HCCV (Chennai, IN)
  • 2019
    Graduated B.Tech Mechatronics at SRM Institute of Science & Technology
  • 2022
    Graduated MSc. Artificial Intelligence at University of Groningen

Projects

Modelling Civil Violence using Multi Agent Simulation

The project is a part of the course "Design of Multi-agent systems". The citizen agents are modelled based on their wealth and hardship. The grievances of the agents are calculated using cop agents and the change in system.


Source Code

Continuous Control

Actor critic training is used to train multiple agents with the goal of balancing objects. The agents share knowledge and optimize actions using a shared experience replay with the aim of achieving higher rewards.


Source Code

Splendor Game with Dynamic Epistemic Logic

The game is designed using PyGame for the course "Logical Aspects of Multi-Agent Systems" and analyzed using Dynamic Epistemic Logic.


Source Code

Multi Agent Tennis

In this project, two agents control rackets to bounce a ball over a net. If an agent hits the ball over the net, it receives a reward of +0.1. If an agent lets a ball hit the ground or hits the ball out of bounds, it receives a reward of -0.01. Thus, the goal of each agent is to keep the ball in play.


Source Code

RL: Navigation

The project consists of an environment with an agent. A reward of +1 is provided for collecting a yellow banana, and a reward of -1 is provided for collecting a blue banana. Thus, the goal of the agent is to collect as many yellow bananas as possible while avoiding blue bananas.


Source Code

Organoid Segmentation using Self-Supervised Learning

Organoids are in vitro cellular structures grown in artificial environments. Self-supervised learning method DINO is used to train teacher and student networks using unlabelled data. A comparison between Vision Transformer and CNN is performed for final segmentation.


Source Code

Certifications

Contact Me

wishalsreenivasan@gmail.com

+91 8148095499