Projects

01
LangGraph System Design React Graph Databses Claude API Data Modeling

Deloitte-Mentored Capstone

Personalized Supplement Recommender

81% accuracy
330K graph nodes
3.4M relationships

An AI system that analyzes health data to generate safe, personalized supplement recommendations grounded in medical evidence.

Built a knowledge graph-driven AI system for supplement recommendations using Neo4j and LLMs - enabling safe, evidence-based reasoning at scale.
Engineered a modular multi-agent pipeline for extraction, normalization, and retrieval with Cypher query validation; deployed via React + Vercel.
Implemented JSON logging and guardrails to enforce auditability across recommendation paths.
View on GitHub ↗ View Website ↗ View Live Demo ↗
02
Prompt Engineering Black-Box Evaluation Benchmarking (AdvBench) Qwen LLMs

AI SAFETY · ADVERSARIAL ML

Jailbreaking Large Language Model

78% success rate
6.2 queries-to-success
25 adversarial objectives

A system that tests how and when language models break under adversarial prompting.

Built an iterative PAIR (Prompt Automatic Iterative Refinement) pipeline to probe LLM safety through attacker–target model interactions
Evaluated Llama-based models under adversarial pressure, measuring success rates and query efficiency across structured attack scenarios
Designed a feedback-driven refinement loop to adapt prompts based on prior responses and improve attack effectiveness.
Closed Source
03
Transformer Architecture Tokenization Transfer Learning Streamlit Text Classification

NLP · Classification

MBTI Personality Classifier

94.8% train accuracy
87% validation accuracy
+12% F1 improvement

A natural language processing system that predicts personality type from text.

Fine-tuned a BERT-based NLP classifier with Hugging Face and TensorFlow - improved F1-score by 12% through optimized tokenization and hyperparameter tuning.
Built an interactive Streamlit app serving real-time personality predictions across 1K+ user inputs with strong generalization on unseen data.
View on GitHub ↗
04
Cross-Validation Model Evaluation Feature Engineering Scikit-Learn

Machine Learning

Credit Card Fraud Detection

0.99 ROC-AUC
0.93 PR-AUC
+18% recall boost

A machine learning system that identifies fraudulent transactions in financial data.

Processed 284K anonymized transactions with a 0.172% fraud rate - addressed severe class imbalance using SMOTE and stratified sampling.
Trained and cross-validated Logistic Regression, Random Forest, XGBoost, and Isolation Forest models with rigorous evaluation.
Optimized decision thresholds to reduce false negatives and boost recall by 18% - where misses carry real cost.
View on GitHub ↗