Education
Brown University
GPA: 4.0
Providence
Sep 2025—May 2026
Pursuing
Master of Science in Computer Science
Brown University
GPA: 3.85
Providence
Sep 2021—May 2025
Degree
Bachelor of Science in Computer Science
Courses
- Data Structures & Algorithms
- Discrete Structures/Probability
- Computer Systems
- User Interface User Experience
- Software Engineering
- Deep Learning
- Cybersecurity Ethics
- Design/Analysis of Algorithms
- Logic for Systems
- Data Science
- Essential Statistics
- Linear Algebra
- Computational Linguistics
- Machine Learning
- Applied Cryptography
- Database Management Systems
- Design + Implement Prog Langs
Proficient in
- Python
- Java
- JavaScript
- PyTorch
- NumPy
- Technical Writing
- TensorFlow
- SQLite
- React
Interests
- Educational Podcasts
- Calisthenics
- Running
Experience
Brown’s Conversational AI Lab
Research Assistant
Providence
May 2024—February 2025
- Built DRIVE, a computer vision model improving CLIP-based relationship detection by 33.5% (text-to-image) and 24.8% (image-to-text), surpassing SOTA models like EVA-02 and SigLIP
- Engineered custom contrastive loss function for CLIP architecture significantly improving image-to-text performance while reducing computational costs across model comparisons
- Created CROCO and CROCO-D datasets for research advancement in relation inference tasks, focusing on distinguishing relational nuances ("fixes" vs "rides") and directional relationships; research currently under publication
Brown University – Department of Computer Science
Teaching Assistant
Providence, RI
Jan 2023 — Dec 2025
- Served as a TA for five semesters across core CS courses:
Data Structures & Algorithms (2x), Data Science (1x), and Machine Learning (2x), supporting over 400 students through labs, office hours, and EdStem discussions.
- Led the Mosaic+ Summer Transition Program as Head TA, designing and coordinating a 4-week Python pre-orientation for 25+ incoming underrepresented students and managing a team of 4 TAs.
- Redesigned and improved course assignments for clarity, rigor, and engagement; delivered detailed code reviews for student Python and Java submissions.
Projects
Secure Multi-Candidate Voting System
Applied Cryptography, Zero-Knowledge Proofs
Systems & Security
April 2025—May 2025
- Generalized an ElGamal-based voting protocol from a 2-candidate design to full k-of-t elections by encoding ballots as vectors of encrypted bits, enabling per-candidate privacy and homomorphic tallying.
- Developed zero-knowledge proofs—including a sum-constraint proof enforcing exactly k selections and correctness proofs for threshold decryption shares—to guarantee ballot validity, verifiability, and secure partial decryption.
- Revamped adjudication and verification pipelines to operate over ciphertext vectors and performed end-to-end benchmarks demonstrating correct tallying, robustness, and strong voter-privacy preservation (evaluated on t=5, k=3).
Reddit Depression Detection
Python, Pytorch, Panda
Personal
Dec 2024—Jan 2025
- Reproduced the approach from "Detecting Symptoms of Depression on Reddit" paper utilizing Random Forest classifiers and achieving evaluation through 5-fold cross-validation and ROC-AUC scoring
- Engineered dual feature extraction approaches combining Latent Dirichlet Allocation (LDA) topic modeling and DistilRoBERTa embeddings from transformer layer 5 to capture linguistic patterns
- Built binary classification system comparing 13 distinct depression symptoms against control posts, with careful consideration for class balancing and temporal separation of control data
BERT Question-Answering System
Python, Pytorch, Cuda, NLP
Comp. Linguistics
Nov 2024—Dec 2024
- Reimplemented the "BERT for Question Answering" paper, surpassing the baseline metrics on short answer dev performance (P: 61.3% vs 59.5%, R: 47.9% vs 47.3%, F1: 53.8% vs 52.7%)
- Used Weights & Biases (WandB) for hyperparameter optimization and experiment tracking, achieving superior performance with minimal training epochs
- Implemented distributed training across multiple GPUs using PyTorch's DistributedDataParallel, reducing training time from 40 minutes to under 15 minutes
Craigslist Item Price Predictor
Python, ML, Web Scraping
Data Science
Apr 2024—May 2024
- Developed a machine learning pipeline to predict second-hand item prices on Craigslist, leveraging VGG16 for image processing and spaCy for text embeddings, resulting in a model that explains 35% of price variance across 3,200+ listings
- Assisted in building and deploying a comprehensive web scraping system to collect Providence Craigslist data, including images, titles, prices, and location data across multiple categories over a 4-day period
- Improved model performance by 27% through implementation of Gradient Boosting Regression and strategic incorporation of location data, reducing median RMSE from 262.70 to 0.99
Portfolio Website
UI/UX, React
UI & UX
Sep 2022—Dec 2022
- Designed and developed a portfolio website showcasing my skills and projects, implementing responsive layouts optimized for both desktop and mobile platforms
- Conducted user research and A/B testing with 10+ participants to gather feedback on navigation and layout choices, iteratively improving the design based on user insight
- Implemented accessibility features including ARIA labels, semantic HTML, and keyboard navigation, achieving WCAG 2.1 AA compliance standards for inclusive user experience
Evolution
Java, OOP, JavaFX
Intro to OOP & CS
Oct 2021—Dec 2021
- Engineered a Java-based arcade system implementing three classic games (Flappy Bird, Snake, Tetris) utilizing object-oriented principles for modular game architecture
- Implemented an AI learning mode using reinforcement learning algorithms, enabling the Flappy Bird agent to progressively improve performance through gameplay
- Designed extensible class hierarchies and design patterns to manage game states, collision detection, and scoring systems across multiple game modes