Vishal Sharma

Software Engineer

About Me

Hi, my name’s Vishal and I’m senior at BML Munjal University. I have experience in Machine Learning, Problem Solving, Algorithms, Data Analytics, etc. I have worked at a number of companies gaining teamwork and leadership skills.

I have worked with a range of different languages and frameworks and I am always looking for opportunities to work with something new. The things that I currently have the most experience working with are: Python , C++ , Java , git , Docker.

I am looking for opportunities that allow me to apply Computational Thinking and Algorithmic problem solving skills to help solve problems that make impact on a big scale and further enhance my knowledge.

Experience

ePlane.ai

theeplanecompany.com

Deep Learning Intern

May 2019 - July 2019

  • Here, I worked as a deep learning engineer and my project was to build Deep Learning models which can enable Autonomous Drones.
  • I learnt Reinforcement Learning and worked on Reinforcement Learning model which could fly Drones in simulations.
  • I also worked on implementation of state of the art deep learning research papers for localization of Drones.

Balnc Care Ltd.

balnccare.com

Machine Learning Intern

Feb 2019 - Apr 2019

  • Here, I worked on a Machine Learning model of Pose Estimation and prediction of future poses.

Education

BML Munjal University

BTech Computer Science

2017 - 2021

  • During my time at BMU I learnt most of my key skills such as team work, project management and clearing strict deadlines.
  • Although we mainly focused on software development and the principles of programming I use a lot of these skills to my advantage in my day to day work.
  • I thouroughly enjoyed my time as university and learnt a lot about a healthy work life balance.
  • I spent a lot of my free time as a committee member of the ACM Student Chapter taking on roles such as Executive Member and Event Organiser.

Projects

Encrypted Nueral Networks

Homomorphic Encryption for Training Nueral Networks

  • This project was part of Cryptography Course at university
  • Trained a Neural Network on Encrypted Data
  • Used Somewhat Homomorphic Encryption system proposed by Zhou and Wornell.

Research Assist

Assistant for Researchers

  • This tool helps researchers to find research papers on topic which they are working on.
  • The researcher provides a link to a sample research paper (pdf) and then this tool parse the text from pdf to create a query which is used to crawl the internet.
  • This tool uses focused crawling which helped reduce search time and improved efficiency.
  • As this tool crawls the internet to search for research papers so it can only find openly accessible research papers.
  • Implemented using Python and its libraries like BeautifulSoup , NLTK , request , PyPDF.

Worked on backend of these tools

  • As a part of Crio Launch program I worked on backend of these applications.
    • QBox: QBox is a secure file-sharing service. It is a customized version of the popularly available VSFTPD server.
    • QMoney: QMoney is a visual stock portfolio analyzer. It helps portfolio managers make trade recommendations for their clients.
    • QCharm: QCharm is an online IDE. Programmers use QCharm to write source code and execute it inside the editor.
  • While working on these projects I acquired various skills like Shell Scriptins, REST API, Code Refactoring, Applied DS, etc.
  • More details of the project can be found in the portfolio issued by Crio.do.

Food Detection

Food Detection with labeled bounding boxes

  • Completed this project as part of hiring task given by a company.
  • This model was able to identify 100 different food items of UEC food dataset.
  • I used Yolov2 with 24 CNN layers.

Sentiment Analysis of movie reviews

Movie Review Classification

  • I worked on this project while I was learning RNN and LSTM.
  • It is a review classification model built using LSTM layers in RNN using PyTorch.
  • This model was able to achieve 83% accuracy for review classification.

Share price prediction

Study and Prediction of share prices

  • I did this project as a part of probability and stats course.
  • I used various Python libraries like NumPy , Matplotlib , Pandas.
  • For prediction I used Monte Carlo Simulations over Geometric Brownian Motion formula used for share prices.

ADrOnest

A Cargo Drone for Medical Emergencies.

  • ADrOnest was built from aluminium scrap with the objective to carry First Aid Kit in case for emergency.
  • Payload capacity of ADrOne was 500gm. We manufactured two versions of the ADrOne.
  • Initial version was made of Scrap Plywood and final version was made of aluminium scrap and Ionised aluminium.
  • This team project was a part of Joy of Engineer Course.

A Little More About Me

Achievements

  • Won College SIH Hackathon and Qualified for SIH Finals
  • Qualified for Crio Launch program (Among top 300/10000)
  • ACM ICPC Kanpur Regional Round(37/106)
  • ACM ICPC Qualifier Round(179/4401)
  • Alibaba Tianchi NLP Hackathon(1/40)

Alongside my interests in networks and software engineering some of my other interests and hobbies are:

  • Reading books (Non-Fiction) and Listening Podcasts
  • Recently started writing on my blog
  • Networking with Professions and People
  • Playing Basketball and Badminton
  • Meditation