Download PDF


"Knowledge, smart work and integrity have no substitute."

Career Plan

I look forward for opportunities to serve in a professionally driven business or research environment where I can explore true potential of my capabilities and contribute lucid and smart solutions to challenging real world problems through data driven techniques, thereby, expand my professional expertise. 

Work History

15 May 201721 July 2017

Summer Research Intern

Data Science Lab, School of Computer and Communication Sciences, École polytechnique fédérale de Lausanne (EPFL), Switzerland

Having been selected in prestigious [email protected] 2017 program among thousands of students worldwide, I worked on a project titled - "Supervised Q-walk for Learning Vector Representation of Nodes in Networks" in the realm of network analysis which involved developing a method for learning vector representation of nodes in a network. The method was inspired by a recently proposed node2vec framework. The latter framework is unsupervised, whereas my approach was a supervised adaptation leveraging label information for improving upon the quality of the learned representation of nodes. The project was developed using Python 3.5,  NetworkX, Numpy, Matplotlib, Scikit-learn, Gensim on Intel Xeon powered Ubuntu 16.04. 

This project was done individually as part of  [email protected]EPFL 2017 programme.

"...Throughout his internship, Naimish put an extraordinary amount of passion and hard work
into his project. He ran extensive experiments and regularly provided detailed updates to
me. Finally, he compiled his findings in a thorough report. In the process, Naimish has
broadened his machine learning expertise significantly through self-directed study.

I am confident Naimish will fare very well in the remainder of his Master’s degree and wish
him the very best for his future career. "

- Prof Robert West, Assistant Professor, Data Science Lab, School of Computer and Communication Sciences, EPFL

Letter of Appraisal can be seen at

The pre-print of the paper is available at: It has been reviewed and accepted for presentation at the IEEE Technically Sponsored Intelligent Systems Conference (IntelliSys) 2018, to be held from 6-7 September 2018 in London, United Kingdom.

22 May 201620 July 2016

Studentische Hilfskräfte (Student Assistant)

Department of Analytical Information Systems and Business Intelligence, Universität Paderborn, Germany

I worked on a project titled - "Facial Key Points Detection using Deep Convolutional Neural Network - NaimishNet".  With my deep learning model, I could secure 18th rank on Kaggle leaderboard in Facial Key Points Detection Challenge.  The project deployed scientific python with deep learning libraries like keras with theano backend.

"...To best of my knowledge and judgement, Mr. Agarwal's work not only constitutes not only a mere implementation but original research advancing the state of the art.

I'd wholeheartedly like to thank Mr. Agarwal for his dedication, skill and hard work. Given the results of his work, I have offered Mr. Agarwal a full semester internship in my lab."

- Jun. Prof Dr Artus Krohn-Grimberghe, Assistant Professor, AIS/BI, University of  Paderborn

Certificate of Completion can be seen at

The project report can be seen at

The pre-print of the paper is available at:


January 2018June 2018

Deep Semantic Segmentation for potential applications in self-driving cars

This project was done individually in 10th semester of IIIT-Allahabad as Masters Thesis work.  The project was done using Python 3.6, PyTorch 0.3 and other libraries in scientific python stack on Ubuntu 16.04.

Project Summary:

Semantic segmentation for low resolution images is a challenging task because the low resolution images lack scene details. The task becomes even more challenging when the number of labelled images are less (~3000 images). Efforts were made to address this challenge in this work by developing two novel models - Hybrid Semantic Segmentation Network (HSSNet) and Full Resolution Densely Connected Network (FRDCN). The two models are developed while keeping in mind the requirement of computational and memory efficiency as needed in present day robotics.
Through a number of experiments,it was showcased that our models outsmart other two popular models - SegNet and LinkNet on Cityscapes dataset for finely labelled low resolution images. We have also compared our models for moderate resolution images with one another. 

August 2017November 2017

Inverse Kinematics Solutions of Robot Manipulators using Meta-heuristic Optimization

The project was done individually in 9th semester of IIIT-Allahabad. The project involves deploying ten variants of differential evolution for computing the inverse kinematics solutions for PUMA 560 robot manipulator. The variants were compared on the basis of their precision and time consumption. The project was done using Python 3.6, PyGMO, numpy, numba, seaborn, matplotlib and pandas on Ubuntu 16.04.

March 2017April 2017

Graph Analysis of Countries and their Official Languages

The project was done as part of the course on Graph Theory in 7th Semester of IIIT-Allahabad. The project used Python 3.5, NetworkX, Gephi, Neo4j and Bash Script on Ubuntu 16.04. 

The project report can be seen at

December 2016December 2016

Signatures Features Visualisation using Deep Convolutional Autoencoder

The project was done as a skill assessment test for selection at EPFL Switzerland for Summer Internship in 2017. Based on my work, I have been selected at EPFL for the duration of 15th May 2017 to 22nd July 2017.

The code repository is located at

February 2016May 2016

Face Recognition using Eigenfaces, Fisherfaces and Support Vector Machines

This project was developed, with 5 team members, as part of Image and Vision Processing course in 6th semester of IIIT-Allahabad. The project deployed R programming language and scientific python with the usage of scikit-learn library for machine learning.

February 2016May 2016

Speaker Recognition using Convolutional Deep Belief Networks

The project was developed, with 5 team members, as a semester project in 6th semester of IIIT-Allahabad. The project deployed the following technologies:

  • Scientific Python stack like Python 2.7, Theano, Numpy, Matplotlib, etc
  • CUDA with CuDNNv4
August 2015December 2015

Static Hand Gesture Recognition using Reinforcement Learning

The project was developed, with 5 team members, as a semester project in 5th semester of IIIT-Allahabad. The project was deployed in pure python.

March 2015April 2015

Tender Management System

The project was developed, with 4 team members, as part of Database Systems Course in 4th Semester of IIIT-Allahabad. The project featured a TMS system similar to present-day Public Works Department TMS. The project deployed the following technologies:

  • Web Development Languages - HTML, CSS, Javascript, JQuery, PHP
  • MySQL Database
November 2014December 2014

Operating System Scheduling Algorithms Simulator

The project was developed, with 5 team members, as part of Operating Systems Course in 3rd Semester of IIIT-Allahabad. The project simulated various OS Scheduling Algorithms like Round Robin, SJF, etc. The project deployed following technologies:

  • Java SE - mainly Java Swing
  • Text to Speech Synthesis Library - provided better Human-Computer Interaction
May 2014June 2014

Airlines Reservation System

The project was developed at HP Educational Services, Kanpur, India as part of the course Java 2 Enterprise Edition Struts with Hibernate Framework. It featured a simpler version of modern-day Airline Reservation Systems. The project deployed the following technologies: 

  • Java 2 Enterprise Edition - Java Server Pages and Servlets
  • MySQL Database
  • Glassfish Server
  • Adobe Dreamweaver 30 days Trial Software

Technical Skills

Machine Learning and Data Science

I have been working in the field of Machine Learning and Data Science since June 2015. Since then I have engaged myself in various projects involving it. I also work with Deep Learning  techniques like Feed Forward Neural Networks, Convolutional Neural Networks, Auto-encoders, etc. 


I have been using Python since October 2015. Since then I have taken online training on various python related topics and also have put it into use in many course assignments and projects. I mainly work with Scientific Python stack comprising Python 3.6, Numpy, Matplotlib, Scipy, Pandas, Scikit-learn, Keras, PyTorch, Pandas, NetworkX, etc.

R Programming

I have used R extensively from June 2015 to May 2016. I have also taken various online courses on Data Camp.


I have used Java extensively in 2009-2010 and 2013-2014 for various course assignments and projects.


I have worked with C# in 2015 since it has a beautiful syntax and at the same time the Visual Studio IDE made it a painless experience to work with C#.

Transact SQL

I have worked extensively with Transact SQL in 2014 only.


I have worked with Gephi in March - April 2017.


I have beginner level understanding of C++. I have used it as part of some course assignments.


I have worked with Neo4j in March - April 2017.

Web Design

Basic web development skills - HTML and CSS

Soft Skills

Strong Work Ethic

I do my job with utmost dedication with full focus on achieving the objective irrespective of what it takes. 

Positive Attitude

have  a positive attitude towards my  work. If some approach fails, I become more cheerful, because I believe that a better approach is yet to be developed. This keeps my motivation level up.

Good Communication Skills

Over the years I have developed good communication skills. I can explain the concepts to a range of people from non-technical to researchers. I can make the concepts really simple and intuitive for non-technical people and full of technical jargon for researchers.

I have sound presentation giving skills. I believe in keeping the content minimal, elegant, complete and sleep-free. 

Time Management Abilities

I can prioritise the tasks and manage multiple activities side by side. In college, we have to do this often in higher semesters. 

Problem Solving Skills

I give my best in solving the problems I face in my work. If I don't know the solution to the problem, I keep meditating on it for hours, days, weeks, etc until I reach a satisfactory solution. Also, when needed, I look for resources online and talk to people about their opinion.

Acting as a Team Player

I like working in teams.  I have acted as the team leader on many occasions in college in which I have done the following:

  • keeping the work going irrespective of difficulties by motivating my team members
  • guiding the team members about the best practices of the work
  • guiding team members  how to critically analyse their work
  • coordinating the work timings for different members with different daily schedules 

As a team member, I have put my best efforts to achieve the target.


I take up the task in which I am confident. If I am asked to do a task about  which I don't know anything, then I do extra labour to acquire sufficient grasp on the domain knowledge until I can say - "yes, I can do it". 

Ability to Accept and Learn from Criticism

I am open to learning and more importantly failures. Failures teach us more than successes. I am always thankful to those people who can criticise my work. Such people are real well wishers since their feedback is crucial in advancing ones performance and work.


I easily adapt to new ideas and changes. I believe that such changes are crucial to the advancement of the abilities of an individual. Fresh approaches to problems kindle one's thought process by providing an altogether new perspective.  I like learning new tools of the trade, reading about different sets of domains, etc. I am always on the lookout of modern best practices so that I can incorporate that into my work.

Working Well Under Pressure

I have past experience of working under pressure with full ability and least mental pressure. I have faced it on various occasions in college while working on semester long projects, exam times, and in internships where I had to give a desired result within a limited time.


July 2013June 2018

5 year Dual Degree programme with B.Tech in Information Technology and M.Tech in Information Technology with specialization in Robotics

Indian Institute of Information Technology, Jhalwa, Allahabad, UP, India


Grade Point Average

1 8.91
2 9.57
3 9.88
4 8.65
5 9.14
6 9.58
Summer Project 2016 10
7 9.89
8 9.50
Summer Project 2017 10
9 9.80


    July 2016December 2016

    Member of Dean's Merit List

    IIIT-Allahabad has listed me in Dean's Merit List for extraordinary performance in 7th Semester. The recognition is given only to the batch toppers. I was among the first few students to have received this award in the same year in which it was started i.e. 2016.

    January 2016June 2016

    Institute Performance Award

    Received a cash prize of INR 36,000 for excellent performance in 6th semester of IIIT-Allahabad. This award is given only to the batch toppers in higher semesters.

    January 2014June 2014

    Merit Incentive Award

    Received a cash prize of INR 36,000 for excellent performance in 2nd semester of IIIT-Allahabad. This award is given only to the batch toppers in lower semesters.

    April 2011April 2013

    School Scholarships

    Based on excellent performance in classes 11th and 12th, I got 100% free education in school. 

    Online Courses Certifications

    Jan 2016Present

    Database Fundamentals

    Microsoft Virtual Academy
    Jan 2016Present

    Visual Studio Code


    Dec 2015Present

    Intro to Statistics with R: Introduction


    Dec 2015Present

    Importing Data into R

    Dec 2015Present

    C# Fundamentals with Visual Studio 2015


    Aug 2015Present

    JSON Data

    Stanford Online
    Aug 2015Present

    Intermediate R


    Aug 2015Present

    Big Data - Fundamentals

    Big Data University 

    Jul 2015Present

    Introduction to R


    Jul 2015Present

    How to work with Quandl in R


    Jul 2015Present

    Reporting with R Markdown


    Jul 2015Present

    Data Visualization in R with ggvis


    Jul 2015Present

    Kaggle R Tutorial on Machine Learning


    Jul 2015Present

    Data Manipulation in R with dplyr


    March 2015Present

    Programming in C# - Jump Start

    Microsoft Virtual Academy
    Jan 2015Present

    Quick Start Challenge - Universal App

    Microsoft Virtual Academy

    Online Courses Certificates Portfolio