Software Engineer with 3+ years building production-grade backend systems, specializing in Java/Spring, cloud infrastructure, and agentic AI workflows.
Software engineer with a proven track record in high-scale backend systems. I've shipped production code at TikTok, State Farm, and COUNTRY Financial. My research on Neural Code Search is published on arXiv.
Building AI-powered developer tools and distributed systems.
From published AI research to interactive visualizers — a few things I've built.
Published research presenting an end-to-end unified architecture for neural code search. Achieved 95% zero-shot accuracy, outperforming CodeBERT, GPT, and T5 baselines.
Full-stack web application enabling students to manage and rate IT courses at Illinois State University. Features user auth, course ratings, and real-time updates.
Interactive visualization of A* and Dijkstra's pathfinding algorithms — real-time grid canvas with drag-to-place walls and animated search.
Animated visualization of sorting algorithms including Merge Sort, Quick Sort, Bubble Sort with adjustable speed and array size.
A fast-paced arcade space shooter game built with Pygame. Features enemies, scoring, and collision physics.
Unbeatable Tic-Tac-Toe AI built using the Minimax algorithm with alpha-beta pruning. The AI evaluates every possible game state to guarantee optimal play.
Visual Sudoku solver using a constraint-satisfaction backtracking algorithm. Watch as it recursively fills the board in real time.
Original unified architecture design for Neural Code Search. Mapped diverse tokenization strategies to custom label representations, prevented catastrophic forgetting via data memory representations for seq2seq models, and achieved 95% zero-shot accuracy — exceeding CodeBERT, GPT, and T5 baselines.
View on arXiv ↗