Hi, my name is
Varun Gangavarapu. I like to build things.
I am a software engineer with multiple years of experience with technologies such as Java, Python, and the web-development toolstack. I have a strong foundation in Data Management and I am skilled in creating responsive full stack applications using React and its ecosystem. Learn More
1. About Me
From my high school days, teaching computer science to more than 50 kids, I was captivated by the world of computer science. I saw it as a race, a thrilling pursuit where every solution led to a new, more complex challenge.
At Rutgers University my passion intensified as, I dove into my studies and personal projects, realizing that striving for the ideal software was a marathon, demanding endurance, adaptability, and an insatiable quest for knowledge.
The next milestone in my journey was an internship at Amneal Pharmaceuticals. where I navigated the complexities of a large corporation, working on analytics for large datasets and making tangible impacts. This real-world experience reaffirmed how far I had come in my race, while reminding me of the journey ahead.
Looking ahead, I know the journey doesn't end. The finish line is ever-evolving, the challenges continue to transform, and my appetite for learning and growth amplifies. I eagerly anticipate the future, with all its opportunities for the next thrilling leg of this ongoing expedition.
Here are a few technologies I have been working with recently:
- JavaScript (ES6+)
- React.js
- AWS
- Java
- Python
- SQL
- Express.js
- Go
2. Where I have worked
- Amneal Pharmacuticals
- Quantitative Finance Club
- Computer Science Remastered
- BluePrint
- Engineering Club
- A2Z Mentoring
Software Engineer Intern @Amneal Pharmacuticals
June 2023 - August 2023
Migrated 100,000s of data points from AWS data lake to modify a visualization layer of the supply chain pipeline to optimize inventory efficiency.
Implemented statistical models such as ARIMA, Exponential Smoothing and Time Series Regression using API’s like Pandas and NumPy in Python to improve demand forecasting by achieving an accuracy of 82%.
Designed an automated script in Python that identified and terminated unnecessary AWS EC2 server instances, leading to a substantial reduction in operational AWS costs by 12%.
Collaborated with cross-functional teams and executive management understanding business needs to facilitate meetings with Google, Amazon, Microsoft, and more to explore AI and Cloud Computing solutions.
3. Some Things I Have Built
Featured Project
Vector Based Spotify Reccomendation Engine
An innovative music recommendation engine that utilizes vector databases and machine learning to analyze users' emotional responses to song lyrics, offering personalized playlists based on mood and sentiment.
- Python
- Flask
- Pinecone
- Spotify API
- OAuth
- JavaScript
- CSS
Things I Have Tinkered With
Here Is The Archive
Sinusoidal Wave Image Compression Algorithm
Created a origional algorithm trying to compress Image Files by representing Pixels as properties of Waves. By using Foriyer Transforms we can composite the waves and revert them back.
- Python
- NumPy
- Matplotlib
Huffman Coding
Implemented Huffman coding algorithm in Java, optimizing data compression and encoding efficiency for various file types, thereby enhancing understanding of tree data structures and priority queues.
- Java
- Trees
- Priority Queues
Infinity War
Implemented 'Infinity War', a data structures and algorithm project, involving implementation of efficient search and sorting algorithms to simulate solutions to virtual conflicrts.
- Java
- Djakstra's Algorithm
- Adjacency Lists
Bomb Lab
Developed a solution for the Bomb Lab project in a Comp Arch class, by carefully analyzing assembly code commands being sent to the system without triggering the wrong isntructions.
- Assembly
- C
- Bit Manipulation
RU Warehouse
Implemented a hybrid data structure combining hash tables and priority queues to efficiently manage simulated inventory for a merchandising project, optimizing memory utilization and speed of accessibility.
- Java
- Priority Queue
- Hashing
RU Kindergarden
A classroom simulation using data structures, including a Singly Linked List for managing a queue of students, a 2D array for seating arrangements, and a Circular Linked List for a musical chairs game
- Java
- Linked Lists
- 2D Arrays