Computer Scientist with expertise in data science, machine learning, and software development. Skilled in Python, SQL, and deep learning frameworks (PyTorch, TensorFlow), with hands-on experience delivering data-driven solutions that improve decision-making and business outcomes. Passionate about building AI systems that are both impactful and sustainable, with research interests in efficient and ethical artificial intelligence.
EDUCATION
Master's in Computer Science (In Progress) GPA: 3.9
New Jersey Institute of Technology (NJIT) | Focus: Machine Learning, AI, Data Science
Bachelor's in Computer Science GPA: 3.69
Rutgers University, New Brunswick | Graduated with Honors
PROFESSIONAL EXPERIENCE
Graduate Researcher – Deep Learning
New Jersey Institute of Technology
Fall 2025
- Evaluated Swin Transformer architectures for image classification, achieving 93.8%–96.35% accuracy on Oxford-IIIT Pet dataset
- Benchmarked against RegNet CNNs and ViT models, demonstrating 11–16% accuracy improvement with computational efficiency gains
- Published comprehensive research findings on hierarchical vision transformers for fine-grained classification tasks
Data & Marketing Analytics Manager
Verizon / JP Levi
- Analyzed millions of customer records using SQL and Python, improving campaign targeting accuracy by 25%
- Built Power BI dashboards enabling real-time KPI monitoring, accelerating decision-making by 35%
- Developed ML models optimizing campaign targeting, increasing ROI by 18% across multiple channels
- Led cross-functional teams in implementing data-driven marketing strategies resulting in $2M+ revenue growth
Corporate Trainer & Assistant Manager
Legacy Marketing
- Managed team of 10+, driving data-backed improvements that raised productivity by 30%
- Designed customer segmentation strategies, boosting sales conversions by 20%
- Implemented performance tracking systems and analytics dashboards for team optimization
TECHNICAL SKILLS
Programming Languages
AI/ML & Data Science
Tools & Platforms
KEY PROJECTS
Fraud Detection System
Python | LightGBM | SHAP
Built ML pipeline analyzing 6.3M transactions, achieving 99.9% fraud recall and 0.9999 ROC-AUC using advanced feature engineering and SMOTE balancing.
Healthcare Management System
PHP | MySQL | JavaScript
Built full-stack system with 20+ normalized tables, role-based access control, and automated drug interaction checking for 500+ concurrent users.
Swin Transformer Research
PyTorch | Computer Vision
Comprehensive evaluation of hierarchical vision transformers, achieving 96.35% accuracy and demonstrating superiority over CNNs and ViT models.
Calorie Burn Prediction
Python | Ensemble Methods
Developed ensemble model achieving 0.0593 RMSLE on 250K samples using weighted LightGBM–XGBoost–CatBoost blend with metabolic feature engineering.
AI-Powered Game Engine
Python | Game AI
Designed strategic game AI with deduction logic and pathfinding algorithms, achieving 90%+ win rate through advanced knowledge tracking systems.
Image Matching Challenge 2025 (CVPR Workshop)
Python | NumPy | PyTorch | OpenCV | NetworkX
Custom pipeline (no pretrained models/OpenCV extractors) for clustering scenes and estimating relative camera poses from unstructured image collections. Harris keypoints with patch descriptors, cosine-sim matching and mutual checks, RANSAC/Essential Matrix validation, graph-based clustering (connected components), Kaggle-format submission.