About
I am a Full-stack Software Developer, and I develop custom software for MedTech clients at Health Connective. I have a Bachelor's and Master's in Computer Science, and my experience spans diverse projects, including full-stack, UI/UX, Data and AI/ML. Outside my 9-5, I am building AI Agents because agents having autonomy to think, plan and act is kinda cool.
Skills
Tech Stack
- Programming Languages: Python, JavaScript, TypeScript, Node.js, C, Java
- Web Technologies: Django, React, Next, Express, jQuery, Passport, HTML, CSS/SCSS
- Databases: MySQL, PostgreSQL, SQLite, Firebase, MongoDB
- AI/ML and Big Data: PyTorch, TensorFlow, NumPy, Pandas, Apache Spark, Hadoop ecosystem (SparkSQL, MapReduce, HDFS)
- Unit Testing: Jest, Jasmine, behave-django, PyHamcrest
Tools & Technologies
- Docker, AWS, Azure, Datadog, REST, OAuth, HTTP, NPM, Linux/Unix, Git, GitHub CI/CD, Figma, Vercel, Postman, Insomnia, Visual Studio, Cursor
Competencies
- Full Stack Development, Distributed Processing, Linear Optimization, AI/ML, Deep Learning, Object Oriented Programming
Projects
- All
- Full Stack
- UI/UX
- Data
- AI/ML
- OS/Linux
- Cool Stuff
Galactic Journal: Online Blog Application
NodeJS | Express | PostgreSQL | HTML/SCSS
A blog application that lets users create, view, edit and delete i.e., supports CRUD operations for blog posts.
Supports multiple users and implements both local and Google OAuth v2.0 authentication strategies using Passport middleware. Incorporates session cookies.
PageSageXpert: Search Engine Application
Python | Apache Spark | Django |
JavaScript
🥇Outstanding Project Award - Rutgers CS
Designed a distributed system architecture to compute pageRank of 5.3 million Wikipedia pages, ensuring
scalability & fault tolerance through Distributed Processing, achieving 100% in accuracy and
overcoming external memory limitations. Integrated suggested articles, text summarizing using BART algorithm,
subgraphs of search results.
Tradewave: Stock Price Prediction using Deep Learning
Python | yfinance | Tensorflow | sklearn
Employed CNN+LSTM, AR-LSTM models to forecast Netflix stock prices for January 2024 using time series analysis.
Integrated Twitter sentiment analysis for nuanced stock predictions using yfinance library and Twitter API.
Demonstrated real-time model evaluation and improved accuracy with reduced Mean Squared
Error, particularly when incorporating Twitter sentiment during training that gave 98% accuracy
MyBookPal: E-commerce Bookstore
HTML | CSS | JS | Node.js |
MySQL
An online bookstore that facilitates seamless purchasing, borrowing, and auction participation.
Utilizing robust REST APIs, it supports inventory management, sales tracking,
and customer relationships. Designed for administrators, customer representatives, and end-users,
it provides a comprehensive online book solution.
UThreads: User-defined Thread Library
C
Created a user-defined thread library to perform multi-threading
operations - creating, yielding, joining, destroying, mutex-locking, and unlocking
threads. Scheduled the threads using 3 scheduling policies: Round Robin, Pre-emptive SJF
and Multilevel Feedback Queues.
Momentum Methods and Descent Directions
Python
(numpy, matplotlib, math)
A study on optimization techniques of Gradient
Descent with and without Momentum for a specific convex function in consideration:
F(x) = 1/2 x^T Q x − x^T c
(where x and c are real vectors of dimension D, and Q is a real symmetric D x D
matrix)
News Article Classification using Deep NLP Techniques
pandas | Scikit-Learn | Heroku | Flask
Created a website that uses Natural Language Processing to classify news articles into 5 domains: Politics, Sports, Business, Entertainment, Technology.
Given a news article or headline, predicts the domain and dynamically fetches top 5 articles in that domain. (Dataset: BBC News API).
Implemented Naive Bayes Classification with XGBoost that gave 85% accuracy.
ElectriXchange
Java | JDBC | HTML/JSP |
MySQL
An online e-commerce auction site for three categories of electronics
items - Phones, Laptops and Desktops. Allows users to create auctions and participate in
manual and auto-bidding over the listed items. Covers three user bases: Admin, Customer
Representatives, End-users.
UMalloc: User-defined Memory Allocation
C
Created a user-defined library for implementations of the functions
malloc( ) and free( ) for Dynamic Memory Allocation,
using First Fit memory allocation and Segmentation policies. The memory management is
programmned using Doubly Linked List to keep track of the addresses of blocks in both
directions.
News Website
NodeJS | Express | HTML/SCSS | NewsAPI
A dynamic news website implemented utilizing an external API "NewsAPI" to fetch daily headlines and news across various categories.
Node.js and Express are utilized in the backend to seamlessly integrate with the API, with robust REST API endpoints, that respond to HTTP requests
Visualization and Analysis of Depressed Users using Social
Network Analysis Tool
Gephi
Social Network Analysis of Twitter users, with modularity, centrality measures, graph
algorithms and layouts revealing key components, inter-dependencies etc., to better
understand the mental behaviours in our social networks.
CSV to SQLite Processing Pipeline
TypeScript | Node | Streams | knex
A data downloading and processing pipeline to download CSV files and add them to SQLite Database.
Downloads a large .tar.gz file, containing lists of organizations and customers, decompresses and unzips the CSVs within,
parses them, and add them to a SQLite database.
Scraping Web Pages: HTML to JSON
TypeScript | Node | Crawlee | Cheerio
A web scraping tool: crawls a given list of Y Combinator ("YC") company urls, scrapes HTML data on their public pages into pre-defined fields in TS interfaces,
and extracts structured data about each into a JSON file. Uses NPM libraries Cheerio and Crawlee.
Career Conversation Bot: My Digital Twin
Python | OpenAI | Gradio | Pushover
An OpenAI-powered AI Agent that emulates my professional persona, as if talking to my digital self.
Uses custom tools to record user details, log out-of-scope queries, and trigger real-time mobile notifications when tool usage is triggered.
Uses the evaluator-optimizer AI agentic pattern to iteratively improve responses.
Resume
Professional Experience
Web Application Developer
Health Connective
Dec 2024 - Present
- Building scalable healthcare systems for MedTech clients
- Full-stack development, maintaining and developing REST APIs, integrating third-party APIs, and unit testing of software features
- Collaborating with cross-functional teams (design and data) to build robust features, and customers to improve user experience
Research Assistant - Distributed Computing
Rutgers University CS Department- New Brunswick
Sep 2023 - May 2024
- Spearheaded a distributed computing research project, designing a PageRank algorithm for 5.3 million Wikipedia pages using Apache Spark, achieving 100% accuracy through optimized computations.
- Developed a search engine prototype with text summarization and visualized large-scale datasets, leading to the Outstanding Project Award from the CS Department.
Graduate Teaching Assistant
- CS 205: Intro to Discrete Structures I - Conducting in-person recitations for topics in Discrete Mathematics: Turing Machines, Program Correctness, Mathematical Induction, Propositional Logic, Finite State Machines etc
Programmer Analyst Intern
Cognizant Technology Solutions
Mar 2022 - Jun 2022
- Developed user interface elements using JavaScript, implemented responsive design principles using React.js strategically enhancing the end-user experience
- Developed and tested REST APIs for backend server architecture using Node.js and Postman
Education
Master of Science in Computer Science
Rutgers University - New Brunswick
2022 - 2024
- CGPA: 3.8/4.0
- Massive Data, Software Engineering, AI/ML, Deep Learning, Regression Analysis, Optimization Mathematics, Computer Vision
Bachelor of Technology in Computer Science and Engineering
Jawaharlal Nehru Technological University
2018 - 2022
- CGPA: 8.3/10.0
- Object Oriented Progamming, Data Structures, Computer Networks, Operating Systems, Cryptography, Web Technologies, Database Management
Achievements
"Outstanding Project Award" for "PageSageXpert" by Rutgers CS Dept
"IT and Web Development Instructor" for underprivileged high school students
1st Place: SPRINT SRIX Challenge - "Smart Helmets using Arduino UNO"
1st Place (College Level): Hackathon SIH 2020 - End-to-end PHP web portal for farmers to buy and sell produce
Certifications
AZ900: Microsoft Azure Fundamentals
Prompt Engineering for ChatGPt by Vanderbilt University
CCNAv7 Introduction to Networks by Cisco
Introduction to Cybersecurity by Cisco
Agile with Atlassian JIRA by Atlassian
Interests
I love singing 🎤 and playing the keys 🎹. I have been a musician for years. I am a professionally certified Keyboardist till Grade 8 from Trinity College London. I am an amateur guitarist too.
Mentoring and teaching are some of my strongest suits. I've introduced 50+ high school students to web development and mentored 200+ college students as a Teaching Assistant at Rutgers.
I'd love to connect!
I'm always on the lookout to expand my network. Let's talk about programming, music, Fortnite, LOTR and more!
If you're exploring the rise of GenAI and Agentic AI too, do you also think we can apply some of the outcomes of the evolution of Social Media Platforms to help us shape AI for the future?
