Kishaan Jeeveswaran

Deep Learning Researcher at NavInfo (Europe) B.V.
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
NL
Languages
  • Tamil Native or bilingual proficiency
  • English Full professional proficiency
  • Sinhala -
  • Hindi -
  • German Limited working proficiency

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
Join now
You need to have a working account to view this content.
Join now

Credentials

  • Build Intelligent Applications - Specialization Certificate
    Coursera
    Aug, 2017
    - Sep, 2024
  • Applied Machine Learning in Python by University of Michigan
    Coursera
    Jun, 2017
    - Sep, 2024
  • Java Programming and Software Engineering Fundamentals
    Coursera
    Jun, 2017
    - Sep, 2024
  • Introduction to Data Science in Python by University of Michigan
    Coursera
    Apr, 2017
    - Sep, 2024
  • Machine Learning by Stanford University
    Coursera
    Mar, 2017
    - Sep, 2024
  • Object Oriented Programming in Java by University of California, San Diego
    Coursera
    Oct, 2016
    - Sep, 2024

Experience

  • NavInfo (Europe) B.V.
    • Netherlands
    • IT Services and IT Consulting
    • 1 - 100 Employee
    • Deep Learning Researcher
      • Jan 2021 - Present

      Research on different topics such as: • Transformers on vision tasks such as object detection and semantic segmentation • continual learning • monocular 3D detection Research on different topics such as: • Transformers on vision tasks such as object detection and semantic segmentation • continual learning • monocular 3D detection

  • Aptiv
    • Ireland
    • Motor Vehicle Manufacturing
    • 700 & Above Employee
    • Master Thesis Student
      • Dec 2019 - Jul 2020

      This master's thesis project focuses on pedestrian trajectory prediction in the urban scenario. For efficient, safe, and smooth drive of an autonomous vehicle, it has to detect all moving entities around it and successfully predict the future trajectories of those entities. Major focus of this project are:• Most state of the art trajectory prediction models are memory intensive which makes it less suitable for deployment in real autonomous vehicles. A memory efficient and robust model is learned to predict pedestrian trajectories.• Study on improving the performance of the model by using explicit context and making the model risk-aware.• Study of the human trajectory datasets in order to select the best one which is closer to the real world urban scenario and which has all the required features to learn the model.

    • Student Assistant
      • Oct 2019 - Dec 2019

      Preparation for master thesis.

  • Catalysts
    • India
    • Human Resources Services
    • 1 - 100 Employee
    • Intern - Software Development (Computer Vision)
      • Mar 2019 - Aug 2019

      The internship is about furthering software development skills and research on computer vision solutions to different problems in the industry. ● Evaluation of different models for object detection of different entities on handwritten documents ● Recognizing the handwritten IBAN numbers using both traditional computer vision methods (OpenCV) and deep learning (Tensorflow, Keras) ● Annotation and training object detection model for hand luggage detection at airports ● Deep Learning on 3D data and graphs

  • Bonn-Rhein-Sieg University of Applied Sciences
    • Germany
    • Higher Education
    • 300 - 400 Employee
    • Research Assistant
      • Sep 2018 - Mar 2019

      Worked in "E-Assessment" research team focusing on building an auto-grading system for Jupyter notebook based exams.● Creation of assignments for the ’Foundations of Probability Theory and Statistics’ course● Writing test cases to autograde the questions with deterministic answers● Coordination of distribution of assignments, collection of answers, tracking autograded scores and manually graded scores using nbgrader● Collection and preparation of a dataset consisting of questions and answers which could be used in the research of automated short answer grading using machine learning techniques

    • Research and Development Project
      • May 2018 - Jan 2019

      Completed a research and development project required for the completion of the master's degree on 'Evaluation of Active Learning for Short Answer Grading'.● Evaluation of different active learning query strategies, machine learning models, and features on different datasets in the task of predicting the grades for short answer questions● Compilation of an in-house dataset of questions, answers, grades and features and making it publicly available● Incorporation of the best ML model into a web-based GUI which can be used to autograde student answers in the future and making it open source

Education

  • Bonn-Rhein-Sieg University of Applied Sciences
    Master of Science - MS, Autonomous Systems
    2017 - 2019
  • Sardar Vallabhbhai National Institute of Technology
    B.Tech, Electronics and Communication Engineering
    2012 - 2016
  • Achievers Lanka Business School
    B.A, CIMA
    2011 - 2012
  • Hindu College, Colombo
    1997 - 2011

Community

You need to have a working account to view this content. Click here to join now