A quick journey through my professional milestones.
May 2024 - Aug 2024
Software Engineer Intern
Atlassian
Drove automation and optimization efforts of security scanning processes for developer platforms across the Atlassian suite under the Product Security team. Supported the ongoing launch of Rovo, focusing on secure data ingestion, training, and promoting ethical AI practices.
May 2023 - Aug 2023
Software Engineer Intern
Alpha Ori Technologies
Executed proof of concept initiatives for quantified analytics-driven testing of API data within a SaaS system, encompassing end-to-end software development, testing, and deployment to enhance server-side efficiency and customer-centric engineering practices.
Dec 2021 - May 2023
Undergraduate Student Researcher
Georgia Tech VIP Program
Pursued research and development under the Georgia Tech Research Institute to construct an automotive LiDAR system, comprising sensor control, data processing software, sensor fusion algorithms, and perception modules to enable real-time object detection and environment mapping for autonomous driving applications.
May 2022 - Aug 2022
University Shadowship
Microsoft
Engaged with Azure Active Directory Team to gain insights into cloud-based B2B and B2C services, while contributing to software development workflow modeling and agile collaboration within sub-teams under a university-partnered training program.
Aug 2021 - May 2025
B.S. in Computer Science
Georgia Institute of Technology
Specializations in intelligence and human-computer interaction.
Contact Me
Let's chat about anything tech, investing, or entreprenurship...
Spotify Music Reccomendation Model
Neural Networks
The ML-Based Music Genre Predictor project employs machine learning to accurately classify song genres, drawing inspiration from recommendation technologies used by platforms like Spotify and Apple Music. Using Clustering, Bayesian Networks, and Neural Networks with filtering, the project successfully enhances the accuracy of music categorization and recommendation. Click here to learn more.
Algorithm Design:
Random Forest,
Clustering,
Bayesian Networks,
Neural Networks with Principal Component Analysis
Financial Modeling & Quantitative Analysis Bot
Backtesting · Risk Management
This project utilizes Python, APIs, web scraping, and data analysis libraries to extract financial stock data and design modeling and visualization tools. It includes backtesting and value investing strategies guided by loss mitigation KPIs, modeling risk, and automating real-time asset value optimization for informed decision-making and maximizing returns. Click here to learn more.
This project delves into the foundational role of search algorithms in computer science, exploring the intricacies of BFS, UCS, A*, bi-directional, and tri-directional searches. It demonstrates how these methods ensure optimal pathfinding through graph traversal, with A* using heuristics to prioritize more promising paths and bi-directional search reducing the explored area. By combining these techniques, the project showcases how to achieve efficient navigation, applying them to both theoretical problems like the Romania graph and practical scenarios like an actual map of Atlanta.
This project applies K-Means Clustering and Gaussian Mixture Models (GMM) to computer vision and image processing, focusing on the segmentation of 2D and 3D images. K-Means groups similar data points into clusters based on proximity to computed centers, while GMMs use a probabilistic approach to handle overlapping clusters and non-circular shapes. By using these methods, the project effectively segments images by color, optimizing cluster identification and enhancing image contrast.
This project implements Minimax and Alpha-Beta Pruning to optimize strategic decision-making in the game "Impact Isolation." Minimax aims to minimize worst-case losses in a two-player, turn-based game, while Alpha-Beta Pruning reduces the number of nodes evaluated without affecting the outcome, making the process more efficient. The algorithm can predict winning scenarios up to 256 iterations ahead of the current game state on a 7x7 grid, providing a robust approach to challenging gameplay with nearly a 98% win rate.
Tech Stack:
C++
Algorithm Design:
Minimax,
Alpha-Beta Pruning
ASL Video Estimator
Hidden Markov Models
This project utilizes Hidden Markov Models (HMM) and the Viterbi algorithm to recognize American Sign Language (ASL) words from sequences of hand positions. By modeling each word with three hidden states—representing the start, middle, and end of the sequence—the system estimates the most probable word based on observed hand movements. It effectively uses training data to map observed sequences to likely word predictions, even for sequences not previously encountered.
Tech Stack:
Python,
NumPy,
Pandas
Algorithm Design:
Hidden Markov Models,
Viterbi Algorithm,
Sequence Modeling and Probability Estimation
