Experience

Applied AI & ML: Solved unexplored problems, shipped features, and built autonomous AI pipelines handling hundreds of millions of users under the Product Engineering organization.

Co-authored a research paper on LLM interpretability accepted for presentation at ACL 2025 SRW; now cited by AWS AI Labs, ICML spotlight papers, and more. Conducted large-scale experiments on Gemma-2 language models to study feature universality using Sparse Autoencoders (SAEs), SVCCA, and RSA. Developed, optimized, and executed PyTorch-based experiments for feature extraction, similarity analysis, and semantic subspace analysis on NVIDIA A100 GPUs, unveiling 73% mid-layer SVCCA alignment.

Built a Java-based terminal messaging system using sockets and multithreading to enable real-time local network communication. Contributed to backend development in an agile, mentor-led environment while learning Git-based version control and collaboration. Selected for an elite cohort of high school programmers.

Explored AI for the first time in high school! Designed and trained NLP and computer vision models under mentorship from Dartmouth and Stanford researchers. Built a mental health chatbot using rule-based algorithms and recurrent neural networks, developed object detection systems for autonomous vehicles, and studied the basics of transformers. Implemented convolutional neural networks with transfer learning (VGG16 on CIFAR-10), achieving 95% classification accuracy with TensorFlow and Keras.

Relevant Coursework: Software Engineering, Data Structures & Algorithms, Object Oriented Programming & Data Structures, Principles of Programming, Information Assurance, Assembly Language, Programming Languages