Welcome to my personal website showcasing some of my research and development works and interests.

In the domain of Computer Science Research, my broad area of interest lies in the fields of Artificial Intelligence, Machine Learning, Deep Learning and their applications in Computer Vision, Natural Language Processing, Reinforcement Learning and their intersections.

In the domain of development, I find myself interested in writing front-end as well as back-end code towards full-stack development of user applications.

My Education and Industry Work Experience

I am currently pursuing my Masters degree in Machine Learning at Carnegie Mellon University.

Before beginning my graduate studies, I was employed as a Software Development Engineer at Adobe Systems Noida, India working in Experience Cloud as well as Creative Cloud teams.

Prior to this, I obtained a Bachelor’s in Technology (Hons.) from the Indian Institute of Technology, Kharagpur in Computer Science and Engineering in 2017.

IIT Kharagpur Webpage

You can find my Full CV (LONG) here You can find my One Page Resume (SHORT) here

Internship and Part Time Experience

In the summer of 2020, I worked as an ML Research Intern at Qeexo and worked on Real-Time Clustering of Sensor Data on Embedded Devices like Arduino Sensorboard.

I am also employed as a Research Assistant under the guidance of Prof. Beibei Li at CMU, responsible for development and maintenance of an Android/iOS research application platform.

I spent the the summers of 2016 as a research intern with Harvineet Singh at Adobe Big Data Labs (BEL), Adobe Research, Bangalore, India and the summer of 2015 as a research intern with Sameep Mehta at IBM Indian Research Lab (IRL), New Delhi, India

About me

I have always been fascinated by the dynamics of the cognitive aspects of the human spirit and the most powerful machine in the world - the human brain. This causes me to possess an intense yearning to seek my career in such an ever-expanding field of machine learning which certainly would invade deep through human psychological aspects and much-complicated thought processes of the human brain in the future. Due to this fascination, I often go through various blogs and the topics that I find extremely intriguing are ‘Artificial Machine Consciousness’ and ‘Artificial General Intelligence’ which never cease to amaze me.

These dreams though, still have several arguments revolving around the possibility of actualization and the closest we’ve been able to advance so far seems to have been the advent of deep neural network architectures that are doing wonders in almost every aspect of intelligence. My broad area of interest for research thus lies in the fields of Deep Learning and its applications in Reinforcement Learning as well as in Computer Vision, Natural Language Processing and their intersections studied in the form of Multi-modal Machine Learning, a vibrant multi-disciplinary research field which addresses some of the original goals of artificial intelligence by integrating and simultaneously modeling multiple communicative modalities such as language, speech and vision.

During my undergraduate course in Computer Science and Engineering (2013 - 2017) at IIT Kharagpur, I was exposed to many interesting subjects like Machine Learning, Computer Vision, Artificial Intelligence and Natural Language Processing to name a few. Moreover, I successfully managed to consistently place myself among the top 3% in my batch of 110 students. Throughout my coursework, I had the opportunity to work on several research projects.

In the final year, I worked on the project ‘Citation Analysis using Deep Neural Networks’ (July 2016 - April 2017). This project aimed to present a supervised machine learning framework to automatically classify citation function using several shallow and linguistically-inspired features. It was directed towards “Prediction of Citation Function of a specified ‘Cited’ paper in the given ‘Citing’ paper utilizing Deep Supervised Learning”, intending to label and point out the sentiment an author carried while citing a given paper, something which even Google Scholar, the most established product in this arena still happens to lack. I primarily deployed well-known Deep Learning architectures to classify the citation sentiments: RNN-based architectures, LSTMs and Text-CNNs. Working on this project, training and comparing several neural network models in TensorFlow, Theano, Torch, scikit-learn and Caffe offered me with a rich and exhilarating experience in getting closely acquainted with these technologies.

Apart from this, I had been a part of a computer vision project ‘Image Deblurring using Convolutional Neural Networks’ (August - November 2016) which involved deploying Deep CNNs for the purpose of Image Convolution. While working on this project, I learnt that direct application of trivial deep neural networks does not seem to be successful in capturing the characteristics of image degradation owing to factors like camera noise, saturation and image compression and thus I along with my team modified our solution approach by establishing a connection between traditional optimization-based schemes and neural network architectures. This particular project offered me with a rigorous learning experience and made me realize that working in such domains requires one to undergo specialized training, significant hands-on knowledge and practical acquaintance with such models and frameworks to be able to succeed. I got to learn more of what goes into tackling a theoretical problem, understanding ideas that led to previous work and trying to build on those ideas to arrive at a solution to the current problem.

Towards the fulfillment of Term Project requirements of the Information Retrieval course, I worked on the development of Researcher Recommendation System (January - April 2016), aiming at offering suggestions to researchers for best-suited peers to collaborate with, based on the similarity of their interests as well as their previous experience and past academic works and locality factors. The project involved constructing a full-fledged Search and Recommendation Engine for the benefit of the Scientific Research community. In the engine, I deployed clustering techniques on the co-author graph and LDA (Latent Dirichlet Allocation) on the author-keyword database to form a collaboration graph and generated compatibility scores based on target node’s centrality and proximity. This project helped me correlate several theoretical concepts I had been taught in the preceding semester to their utility in applications.

In my final year, I also worked on the task “Deployment of PCF-based models for Performance Enhancement of Double Channel 802.11n WLAN” (July-November 2016) as a Term Project under the Network Modelling course. In case of Wi-Fi enabled hand-held devices, the demand of high throughput introduces high throughput standards such as IEEE 802.11ac which supports Dynamic Bandwidth Channel Access (DBCA), where a wireless station selects channel bandwidth dynamically based on the availability of the secondary channels. In our solution approach, we developed a hybrid adaptive resource reservation mechanism, Hybrid Adaptive DBCA (HA-DBCA), a new algorithm to mitigate the issue of unfair channel-access which arises due to the use of contention-based medium access mechanism in DBCA. Eventually, we improved the work and authored a paper which was accepted and published in one of the reputed conferences, the 42nd Annual IEEE Conference on Local Computer Networks (November 2017). The learning experience was huge. I learned how to devise and build tests to empirically support theoretical findings and about the intricacies of authoring a quality research paper. This successful publication in such an esteemed conference forum significantly lifted my spirits and confidence in my skills and abilities as a researcher.

All my endeavors into the area of Machine Learning have taught me a lot and given me a flavor of contemporary research in various areas within this field. Along with these research projects, I have also worked on several development projects namely 2D/3D Graph Plotter & 2D Graph Scanner Software (2016, 2017), Course Management System (January - April 2016), TinyC Compiler (July - November 2015), Personal Library - User & Issue Management System (January - April 2015) and Software Component Cataloguing System (January - April 2015). These experiences have taught me how large teams collaborate on huge projects with a particular end-user product in focus.

Apart from academics, I have also been quite active in several extra-curricular activities during my undergraduate years. Amongst several student activity societies operational at the campus, I had been a part of K.R.A.I.G. (Kharagpur Robotics and Artificial Intelligence Group), the technical wing of Technology Robotics Society at IIT Kharagpur. I had also been a member of the AUV (Autonomous Underwater Vehicle) group for six months wherein I worked on the software design upgrades for the utility vehicle drone engines. Also, having been appointed as a teaching assistant for a few students during the final year helped me gain quite an insightful understanding of the teacher-learner relationships which I believe aided me to grow into a better learner myself.

Owing to my passion for gaining hands-on and practical experience in the industry, I worked as a research intern with Adobe BEL (Big-Data Experience Lab) Bangalore (May - July 2016). During this period, I worked on the project “Analytics of Multi-Channel Customer Data”, under the guidance of Research Scientist Harvineet Singh from IIT Delhi. We applied a multitude of Machine Learning models to predict customer affinity to interact in a given channel with limited information, given such data of interaction across other channels. Our approach was unique and novel as it suggests ways for predicting customers’ purchase behavior exploiting a shared feature space which captures cross-channel correlations instead of directly merging the input channels’ data or predicting based on single channel. For this, we used CCA (Canonical Correlation Analysis) for projecting the observed input variables to an underlying “latent” or shared space, which had never been applied earlier in the field of customer data analytics. After completing the second year of my undergraduate degree, I also had the opportunity to work as a junior research intern with IBM IRL (India Research Lab), New Delhi (May – June 2015), where I worked extensively on the project “Construction of Knowledge Graph and Use of Semantic Knowledge Graph in Supervised Training and Blended Learning”. Being a part of these internship programs proved to be quite a stimulating exposure and allowed me the opportunity to get acquainted with the finer details of the course one follows while working on industrial research projects.

I was employed as a Member of Technical Staff at Adobe Systems, Noida (June 2017 – June 2019). My roles and responsibilities include software development, improvement and maintenance. At the moment, I work as a part of the product team affiliated to AEM (Adobe Experience Manager) Forms, an element of Adobe Experience Cloud Solutions, developing Intelligent Chatbot services using NLP for Query-based interaction on personalized customer contexts derived from their respective statement reports. Being a naïve professional, each day at the office offers me plenty of experience and learning opportunities as I acquire new skills and learn to deliver, contributing to my team’s goals. Eventually, I now appreciate the additions to my capability such a work experience has offered me with, as I recognize substantial improvements in my teamwork skills and find my project collaborations getting more effective on daily basis. I have also learned how to quickly dive into, understand and maintain large codebases. Apart from my office work, I have also been involved part-time in societal projects, contributing as a freelancer to two live projects at my hometown, Sufi Information Portal, Durg District (December 2017 - April 2018) and a website for automation of activities of Talpuri International Colony Association, Bhilai, Chhattisgarh, India (July - November 2017).

The latter project also recognizes my role as a System Administrator as on date. My active participation in such real-world projects has added to my confidence in my capabilities as an application developer and furthered my skills to practically apply my knowledge in realistic scenarios. Nevertheless, to be able to advance to the next level, incorporating a layer of intelligence in my implementations and to develop smart applications still remains a dream to me which makes me want to go for higher technical education in the aforementioned fields.

Having thus gained plenty of experience in both research and development, the two constitutional aspects of the IT industry, I now believe myself to be qualified and competent enough to master the higher and more advanced spaces of Computer Science.

Post completion of my master’s program, I aspire to put my research and engineering skills to use in the industry by contributing to the real-world research problems as a research engineer in companies leading in technological innovation and shaping the future of AI.

Personal History Statement

I’m an Indian, born and raised in a steel township of Bhilai, Chhattisgarh, India. Bhilai is a renowned place located in the central province of the country, known for its Steel Plant and quality raising of future generation techie kids owing to quality educational environments in the geographic region. Following the globalization in my country, quite a lot has changed especially when we talk about values, morals, traditions, and practices of the people in my nation. I have learned to adapt to these changes while still being in keeping with the Indian roots, culture, and traditions that I have inherited from my ancestors. My town is situated in the middle part of the country and is gifted with the presence of a diverse range of ethnic groups from all parts of the nation: Punjabis, Gujaratis, South Indians, Marathis, Christians, Muslims and many other sects of the Indian nation. Since childhood, I’ve lived among people belonging to these groups in school, college as well as in my neighborhood holding belief in integrity in diversity. It is quite true that despite the community efforts to introduce uniformity among all; there still remain differences among individuals belonging to different ethnic factions, which could easily be perceived by me since early childhood. I nevertheless constantly tried to eliminate such perceptions and encourage uniformity among my mates. By the time I was in middle school, I had learned several skills such as drawing and painting, playing keyboard, skating and swimming. I found immense pleasure and sought perfection in whatever activity I pursued. These experiences provided me with a culturally vibrant atmosphere and opportunities to interact with people belonging to different backgrounds. Furthermore, I have often been elected as representative of my classmates, colleagues or team in school, college and office. I had been chosen as the team captain for my hostel for leading a team of sophomores in the inter-hall software competition. After being promoted to a senior position at the tech-societies at my college, I have guided and mentored many students joining as junior members in the team. I had also been appointed as a teaching assistant and student mentor in the CSE department of my college. In my school years, I played the role of class captain several times and had led a team of thirty in the school choir group. While playing any such roles of leadership, I always try to ensure equal consideration to all members I speak for. My leadership skills were gradually groomed by these experiences, and I also enjoyed serving other students, being able to bring smiles to them and spread happiness. I am also a member of a spiritual-cum-social organization which holds its convictions of the transformation of mankind through love and brotherhood establishing peace and happiness globally. Being a part of this group, we regularly get together to participate in collective-effort activities like “Shram-Daan” (physical contribution to the development of “Aashram”), social-service jobs (donations to less privileged sections) and prayers and “Satsung” (ceremonies in divine company). In middle school, I became fascinated with Science and Mathematics and secured several honors in the school Olympiads. I enjoyed participating in these intellectual challenges and secured top positions in these examinations, bagging accolades and scholarships which aided my educational finance. In addition, my family has always been a middle-class one belonging to average-income groups. My father is a technical graduate from Indian Institute of Technology, Varanasi and working with M/S Steel Authority of India Limited, a government enterprise that has been a strong backbone in the progress of the Nation.

My mother is working in Academia since last twenty years in the capacity as a Professor in Department of Computer Science and engineering of Bhilai Institute of Technology, Durg, Chhattisgarh, the premier most institution of our province. She is a doctorate in the faculty of Computer Science with her specialized areas of research interests in Data Mining and Artificial Intelligence. Moreover, I feel blessed to be a member of a family who had been and even now is involved in professions relevant to educational and scholarly careers. I always worked hard and did my best to avoid the further financial burden of my education on my parents by trying to seek as many scholarships, grants, waivers, remissions and financial awards as possible.

Throughout my education years, I always strived hard to consistently seek perfection in almost every endeavor I pursued which can be demonstrated by the grades I obtained in my undergraduate as well as the fact I topped among my school batch mates in every class. I secured a 100% score in the subjects of Physics, Mathematics, Chemistry as well as Computer Science in my Class XII Board Examinations. (Higher Secondary Education) and had obtained a complete 10.0 CPI in my Class X Board Examinations (Secondary Education).

Also, I was able to bag prestigious national level scholarships like NTSE (National Talent Search Examination) and KVPY (Kishore Vaignaynik Protasahan Yojana) both of which require several tough rounds to be cleared and are quite coveted by our nation’s students.

I chose to remain firm in my decision and remain confident about my intention to learn dynamically changing technologies in the field of Computer Science that I find so fascinating.

My elders have always preached not to waste even the slightest of opportunities offered by some of the life’s moments, which is why they encourage me to study hard and to push beyond my limits and boundaries in order to pursue my dreams. Hence, I believe that my ethnic heritage did not only prepare me well to pursue this degree but has also afforded me the motivation to persevere and to make my parents and grandparents proud who have always been there to guide me through every step.

Letter of Recommendation

It is my pleasure to advocate the application of Varun Rawal, who graduated from Department of Computer Science in 2017, for admission to research fellowship program at your esteemed organization. I am a Professor at the Indian Institute of Technology, Kharagpur and have known Varun since his sixth undergraduate semester, when I was his professor for Networks Theory course (Grade : A) as well as his advisor for the Networks Lab course (Grade : EX) in which he completed all the assignments with superior quality work and performed excellently well in all the evaluations.

His overall intelligence and enthusiasm for computer science are clearly reflected in his grades for almost all the courses, which were by far one of the best in the class. Our university curriculum requirements and expectations are nationally acknowledged for their high standards. Despite such demanding workload, Varun managed to secure a final CGPA of 9.72 / 10.0, which I believe is incredibly hard to score, something one simply cannot achieve without sufficient dedication, hard work and passion for the subjects in Computer Science.

Being an extraordinarily bright student of our department, he was offered two prestigious scholarships from IIT Kharagpur: Gora Lal Syngal memorial Scholarship 2015-16 and B.P. Poddar Scholarship 2015-16, awarded to the best students of the final year class.

His abundant research experience can be illustrated by his various research-oriented pursuits and endeavors: working as a research intern in IBM and Adobe Research labs and participation in several research undertakings as term projects under various courses. In his final year, he expanded on his work in the Network Modelling course in which he along with his team went well beyond the curriculum and successfully published a paper which got accepted in the top IEEE conference on Local Computer Networks (LCN) in 2017.

The primary source of credibility I earn as a recommender for Varun is by the virtue of the time I spent with him at Adobe Big Data Labs, Bangalore where I spent my summer 2016 as the main visiting research advisor. Here, I guided him in capacity of his Research Internship Project Advisor on “Analytics of Multi-Channel Customer Data”, majorly an ML-driven venture solved by deep-neural-networks based techniques. During this time, I took note of several qualities he demonstrated over a span of three months working in collaboration with me. Being the most active among his teammates, his teamwork skills and leadership qualities were conspicuously appreciable throughout the tenure. He took initiative and sought responsibility and displayed an extremely enthusiastic attitude towards research, working on all the relevant research papers with great absorption.

I must corroborate his undoubtedly effective communication skills by my own experience during the discussions we had during the brainstorming sessions where he demonstrated his oral articulateness via clear and concise arguments. I also found him to be highly proficient in applying the theoretical concepts we came across in various research publications to the relevant aspects of our work in the best suitable manner. His presentation during the final assessment of his internship project looked very promising and we recommended he should go ahead and submit his work to a reputed conference. I currently act as the chief head of very active research group : “Complex Networks Research Group” (CNeRG) laboratory at IIT KGP, where he worked in collaboration with PhD Research scholar Mayank Singh under the supervision of Dr. Pawan Goyal on the research project ‘Researcher Recommendation : Relevance’ in the course Information Retrieval, where they exploited a combination of LDA-based techniques and collaboration graph algorithms to come up with a novel approach for building a recommendation engine for the benefit of the scientific community, demonstrating sharp reasoning skills towards solving a challenging problem.

As per my recent discussion with Varun, I found him to be particularly interested in applications of ML algorithms to problems in the field of Vision and Language.

There are a lot of traits that I believe make him stand out of the crowd such as his thirst for science and a constant will to stay at the front and seek perfection in every task he pursues. Considering my classroom experience and interaction with him, I would find no hitch in substantiating his punctuality in regular coursework, ability to pay attention to minute details and the habit of taking every aspect of a task seriously all of which I find to be considerably lacking in majority of today’s generations. In conclusion, I must say that getting admitted into the much sought CSE department at the University of IIT Kharagpur is by its very nature a noteworthy accomplishment amongst Indian engineering aspirants and hence I find every student we deal with to be a brilliant mind in himself. However, Varun has clearly gained a special place in my judgment. I would specifically rank him among the top 1% of students that I have taught in the past few years in respect of his hard work, persistence in efforts and passion for research.

I believe he is a self-motivated student who is well-disciplined and industrious with a pleasant personality. I am very glad he has chosen to attend research fellowship program and decided to pursue a career in research in the field of Computer Science. His initiative and determination make him an ideal candidate and I strongly endorse Varun for any industry or academic institution. He would definitely prove to be an extremely positive asset to any development or research community.

For more info

Reach out to me at Email / LinkedIn / Github.