Harsh Rao Dhanyamraju

About

Learn more about me

Computer Science Graduate

I am an ambitious individual who is self-driven to leave his mark as a 21st century professional. I love challenges and never hesitate to take initiatives. I aspire to learn from experts in the industry and desire to create a unique niche for myself.

University: Nanyang Technological University
Website: HarshRaoD.github.io
City: Singapore
Home Town: Mumbai, India

Age: 22
Degree: B.Eng in Computer Science
Email: harsh.rao@dhanyamraju.in
Hobbies: Chess & Table Tennis

It is my dream to be a part of transformation and innovations led by technology that bring about radical changes in the way organizations and countries operate. I aspire to get experiential learning during my college years in varied areas to be able to take my learnings further.

Months of
Internship Experience

4.5 / 5 CPGA

(Highest Distinction)

Technical
Projects

All India Rank
JEE Main 2020

Top Skills

Python Extremely Familiar

PostgreSQLHighly Familiar

PyTorchFamiliar

Node.jsFamiliar

JavaFamiliar

OpenCV Decent

All Skills

Python

PostgreSQL

NodeJS

Java

HTML

CSS

PyTorch

OpenCV

TypeScript

Flask

Docker

Linux

React

TensorFlow Lite

Supabase

AWS S3

Jenkins

MongoDB

Heroku

Spring Boot

Testimonials

I was the supervisor of Harsh when he was a full-stack developer intern in SAP. He is very knowledgeable and has strong software engineering fundamentals. He is a good communicator and positive teammate who shows great initiative. I recommend Harsh without reservation and strongly believe that he will be a valuable addition to any company that he joins in the future.

Lenald Ng

Software Engineer, SAP

I have watched Harsh develop independent thinking and problem solving skills and demonstrate a strong knowledge of python. I observed that he demonstrated excellent communication and teamwork skills, high learning ability and a very positive work attitude.

Mark Wong

CEO, Occutrack Medical

He is extremely dedicated and self-driven, and I always admire his drive to build things better than before. It was an enlightening experience working on this project. Harsh is a very creative person, and he always thinks outside of the box.

Mandar Thosar

Mentor, X2P-Excel-to-PowerPoint-Convertor

CV

Curriculum Vitae

Internship Experience

SAP (AI Business Services)

Full Stack Developer Intern

Jan 2023 - Jul 2023

Built CI/CD Pipelines using Jenkins and SonarQube
Deployed applications to cloud using Cloud Foundry and Docker
Wrote Unit Test to increase code coverage up to QA standards
Refactored multiple REST APIs to patch security vulnerabilities, reduce code smells, and increase code quality
Liaised with external vendors while organising 2 events: SAP x DataScienceSG and DCOM Singapore
Participated in SCRUM Meetings including daily stand-ups, sprint planning, sprint reviews, and retrospectives

OccuTrack Medical

Machine Learning Intern

May 2022 - Jul 2022

Forecasted Trends using ARIMA, SARIMA, GARCH, Autoformer, and ETSformer
Explored Computer Vision Algorithms including CNNs, ViT, SWIN Transformers, STEGO
Engineered applications with OpenCV, Mediapipe, Dlib, ORB, Yolo, and GANs
Created Interactive dashboards with Plotly, Seaborn, Matplotlib, HTML, and CSS
Trained machine learning algorithms using XGBoost, Scikit-Learn and Statsmodels

Nuvama Wealth & Investment Ltd (then: Edelweiss Broking Ltd)

Intern (Corporate Technology)

May 2021 - Jul 2021

Automated sending of invites for live events using MS Power Automate and SharePoint
Integrated ElastAlert2 with Python, Elasticsearch, and Kibana
Configured Microsoft IIS Server

Education

Bachelor of Engineering (Computer Science)

2020 - 2024

Nanyang Technological University, Singapore

Specialising in Artificial Intelligence and Cybersecurity
Honours (Highest Distinction), CGPA: 4.5 / 5.0
Relevant Courses: (1) Neural Networks & Deep Learning, (2) Natural Language Processing, (3) Machine Learning, (4) Big Data Management, (5) Applied Cryptography, (6) Network Science, (7) Software Security

Co Curricular Activities

Institution of Engineers Singapore, NTU Student Chapter

Honorary General Secretary

Aug 2022 - Aug 2023

Administered the development of IES-NTU’s website
Strategic Oversight on IES-NTU’s operations
Organised a workshop on Coding Interview Preparation with 150+ participants

School Of Computer Science and Engineering, NTU

Student Tutor

Aug 2022 - Nov 2022

Explained the concepts of Probability and Statistics
Encouraged introspection and critical thinking

Institute of Electrical and Electronics Engineers, NTU Student Branch

Tech Officer

Sep 2021 - Jul 2022

Deployed iNTUition’s virtual platform (South-East Asia’s largest student-led hackathon)
Conducted a workshop on Natural Language Processing (Techathelon Series)
Facilitated a workshop on Graphs and Trees (Coding Nights)

Hall of Residence 11

Chess Captain

Aug 2021 - Jul 2022

Led Hall 11’s Chess Team as Captain in Inter Hall Recreational Games (IHRG)
Organised Training sessions, Practise games, and Team Bonding Events

Experience

Internship Experience

Full Stack Developer Intern

During my internship with the Personalisation Recommendations Team, which is a part of AI Business Services, I had the opportunity to work on various tasks. One of my major accomplishments was building CI/CD Pipelines using Jenkins and SonarQube, which streamlined the development and deployment process. Additionally, I gained experience in deploying applications to the cloud using Cloud Foundry and Docker, enhancing the scalability and accessibility of the applications. Another important aspect of my role was ensuring the quality of the codebase. To achieve this, I wrote Unit Tests to increase code coverage up to QA standards, guaranteeing a higher level of reliability and reducing the risk of potential issues. Furthermore, I took the initiative to refactor multiple REST APIs to address security vulnerabilities, eliminate code smells, and ultimately improve the overall code quality of the system.

Machine Learning Intern

During my tenure at Occutrack, I had the opportunity to work on various proof of concepts (POCs) in relation to the development of our upcoming flagship product. These projects involved forecasting trends through the utilization of several advanced techniques such as ARIMA, SARIMA, GARCH, Autoformer, and ETSformer. Additionally, I delved into the realm of Computer Vision Algorithms, exploring cutting-edge models such as CNNs, ViT, SWIN Transformers, and STEGO. To engineer applications with robust capabilities, I harnessed the power of libraries like OpenCV, Mediapipe, Dlib, ORB, and Yolo. This allowed me to develop solutions that leveraged computer vision technologies effectively. Furthermore, I played a pivotal role in creating interactive dashboards using tools like Plotly, Seaborn, Matplotlib, HTML, and CSS, which not only facilitated data visualization but also enhanced the user experience. To ensure optimal performance and predictive power, I undertook the responsibility of training machine learning algorithms utilizing XGBoost, Scikit-Learn, and Statsmodels. This enabled us to build models that could make accurate predictions and provide valuable insights.

Intern (Corporate Technology)

During my internship with the Corporate Technology team at Edelweiss Broking Ltd (Now Nuvama Wealth & Investment Ltd), I had the privilege to contribute to various pivotal projects. I spearheaded the automation of event invite distribution through adept utilization of MS Power Automate and SharePoint, streamlining the live event experience for our stakeholders. Additionally, I assisted with the successful integration of ElastAlert2, harnessing the power of Python, Elasticsearch, and Kibana to enhance the efficiency of our alerting system. Furthermore, I played a key role in configuring the Microsoft IIS Server, bolstering our infrastructure's robustness. These experiences not only honed my technical skills but also provided me with profound insights into corporate technology operations. Grateful for the opportunity, I look forward to translating these learnings into my future endeavors.

Experience

Co Curricular Activities

Honorary General Secretary

During my role as Honorary General Secretary for the Institution of Engineers Singapore (IES) NTU Student Chapter, I provided strategic oversight on the chapter's operations. I successfully spearheaded the subcommittee recruitment initiative, fostering a dedicated and dynamic team. One of my notable achievements was orchestrating a highly attended coding workshop that garnered over 150+ enthusiastic participants, resulting in a full-house event. In addition, I played a pivotal role in administering the development of the chapter's website, enhancing its online presence. My commitment to driving excellence and fostering growth within the engineering community is reflected in these accomplishments.

Student Tutor

During my time as a student tutor teaching probability and statistics, I have gained invaluable experience that has enriched both my academic and personal growth. Guiding my fellow students through the intricate concepts of probability theory and statistical analysis has allowed me to deepen my own understanding of the subject matter. The joy of witnessing my students understand complex ideas has been incredibly rewarding. Moreover, this experience has honed my communication and presentation skills, enabling me to convey complex information in a clear and accessible manner. Interacting with diverse learners has also broadened my perspective and taught me to adapt my teaching approach to cater to individual learning styles.

Tech Officer

In my role as a Tech Officer at the Institution of Electrical and Electronics Engineers (IEEE) NTU Student Branch, I demonstrated a strong technical acumen by successfully debugging the virtual platform for iNTUition, the largest student-led hackathon in South-East Asia. I also had the privilege of imparting knowledge by conducting a workshop on Natural Language Processing, and further developed my organizational skills by facilitating a workshop focused on Graphs and Trees. These experiences reflect my dedication to fostering innovation, sharing expertise, and contributing to the advancement of technology within the IEEE and broader academic community.

Projects

Project Work

Generalisation Capacity of NLVL Models

Github Report Slides

This paper presents a thorough investigation into the Generalization capacities inherent in Natural Language Video Localisation (NLVL) Models. NLVL constitutes a multimodal machine learning endeavor wherein the model, when provided with a video and a textual query, endeavors to identify and retrieve the segment of the video most closely aligned with the query. Our study involves mitigating the temporal distributional bias within the Charades-STA dataset, evaluating a model's ability to generalize across first-person videos, and appraising its robustness against errors and perturbations within the text queries.

Analysis of Deep Image Colourization techniques

Github Report

An Evaluation of various different techniques used to train models for the Image Colourization task. We experiment with different model architectures, loss functions and output formats in an effort to make to convincing coloured images from black and white inputs. Our models are based on Deep Convolutional Neural Networks (CNNs).

PyTorch | OpenCV | Python | Computer Vision | GenAI

PreSchool Go Where

Github Website

An interactive web app, developed leveraging the power of the Google Maps API and data from ECDA, to help parents find the right PreSchool for their child. It allows to parents to browse through several PreSchools in Singapore using a variety of filters like: Location, Second Language, Dietary Preferences, Cost, etc. This is especially helpful to children with special needs. The platform will also send the customer an email of with the contact details of the schools they chose.

Horse Health Prediction

Video

Our solution to a Kaggle competition focused on predicting equine survival outcomes using provided health parameters. Our approach involved rigorous data analysis, feature engineering to enhance predictive accuracy, and implementation of ordinal encoding to streamline the dataset's feature set. Subsequently, we trained various traditional machine learning models, including XGBoost, LightGBM, HistGB, and Random Forest, achieving a commendable placement within the top 6.7% globally on the Kaggle competition's public leaderboard.

Machine Learning | Data Science | Scikit-Learn

Learning With Errors Based Cryptosystems

Slides Github

Designed a Quantum Safe Public Key Cryptosystem based on the Learning with Errors (LWE) mathematical problem and its variant Ring Learning With Errors (RLWE). Investigated the system's vulnerability to the Arora-Ge Algebraic Attack. Evaluated the use of Ring Learning With Errors (RLWE) for Homomorphic Encryption.

Public Key Cryptography | Cybersecurity

Pegasus & HeartBleed

Slides

A computer security case study exploring the workings of the Pegasus Spyware on iOS devices and the HeartBleed vulnerability collated into a 15 min presentation. We explore the series of vulnerabilities exploited by the Pegasus Spyware to gain access to the victims private information. We then explore the exact cause of the HeartBleed vulnerability in OpenSSL and analyse the code to explain the patch.

Cybersecurity | Spyware | iOS | C | OpenSSL

IntelliCheck

Demo Notebook

A Proof of concept (built in 48 hrs as part of the AI Affinity hackathon) for a revolutionary application that uses AI to correct handwritten examinations. The teacher can specify his/her own marking scheme. The software also provides personalised feedback to each student, helping them improve. The AI is built using state of the art handwriting OCR models from HuggingFace Hub and computer vision models from OpenCV for image preprocessing. The AI then uses state of the art NLP models from Flair, NLTK, and SpaCy to verify if the students answer is correct. The POC currently only works on theorical questions with factual answers.

Multi Disciplinary Design Project

Design a robot to complete a maze in the shortest time possible. The maze consists of a a series of obstacles to avoid. Each obstacle has a picture of a certain letter or number attached to it. The robot has to navigate the maze, capture all the images, identify them, and send the results to an android tablet. Using a four wheeled robot, STM Controller, Infrared Sensor, Ultrasound Sensor and Raspberry Pi.

TensorFlow Lite | Object Detection | Roboflow | Computer Vision | Raspberry Pi (Linux)

myBike SG

Website

An interactive web app, developed leveraging the power of the Google Maps API and data from LTA, to help people find safe places to park their bicycles without fear of being fined. The web app helps cyclists find a bicycle rack close to their current location or destination. It also helps them find their way through the city. It can also store new information regarding bicycle racks in Singapore to build on the existing data.

Restaurant Reservation and Point of Sales System

A command line application in java that allows users to reserve tables, take orders, and generates the a bill for the customer and sales report for the client. Includes documentation like UML Class and Sequence Diagrams and Java Docs (in HTML)

Java | UML | Java Docs

Jousting

A fun retro and arcade style game inspired by medival knights. Jousting is a 2 player game where both knights try to poke with a spear. A single player mode with 2 different difficulty levels is also there.

Python | Tkinter | PyGame

Olist Data Analysis

Analysed the Olist dataset on e-commerce spending in Brazil (kaggle). Evaluated different cities based on their spending patterns. Trained a machine learning algorithms using Random Forest and Support Vector Machine

Python | Pandas | Sckit-Learn | Seaborn

X2P - The Excel to PowerPoint Convertor

A one of a kind desktop application that allows users to create their own PowerPoint presentations using graphs created from data stored in Excel, csv, json files. The application also allows the type of graphs to be saved as templates and then reused. This is especially useful for regular presentations like Monthly Sales Reports etc.

Python | Tkinter | Matplotlib | Seaborn