- 🎓 Graduate Student – M.S. Data Science @ University of New Haven (May 2026)
- 🛠 Currently working as an AI Automation Engineer (Extern)
- 👨💻 Experience as Software Developer Intern (Uplaud) and Full Stack Developer Intern (Cognizant)
- 📊 Passionate about data engineering, industry insights, machine learning, economic research, and cloud analytics
- 🚀 I design end-to-end systems: Data pipelines → ML Models → Dashboards → Deployment
- ⚽ Fun Fact: Played college-level soccer; teamwork and discipline are core to how I work
More about how I work (click to expand) ✨
I care deeply about clarity, performance, and creating systems that last. I believe in "measure twice, build once" — prioritizing stability, clean architecture, strong documentation, and thoughtful experimentation. I always bring a mix of creativity & engineering discipline to everything I build.
| 🏫 University | 🎓 Degree | 📊 GPA | 🎓 Graduation | 📘 Relevant Courses |
|---|---|---|---|---|
| University of New Haven | MS in Data Science | 3.5/4.0 | May 2026 | Data Engineering, AI, ML, Deep Learning, NLP |
| Sapthagiri College of Engineering | BE in Information Science | 8.5/10 | May 2023 | DBMS, Operating Systems, DSA, Networking |
| Project | Description | Stack |
|---|---|---|
| 🌆 Urban Car Crash Risk Radar (AWS Data Engineering) | Cloud-native data engineering system built to collect, clean, and model millions of crash records across LA, Houston, Detroit, Dallas, and Memphis. Designed Glue ETL jobs, partitioned Parquet datasets, and optimized Athena queries — reducing query latency by 40–60%. | AWS Glue, Athena, S3, ETL, Parquet, SQL, Data Engineering |
| 🏡 Real Estate Price Predictor (ML Analytics App) | Multi-model ML platform using XGBoost, Random Forest, and Linear Regression to predict house prices with trend insights and region-based analytics. Enhanced with data preprocessing and visual insights. | Python, XGBoost, Random Forest, Linear Regression, Pandas, NumPy, ML |
| 😊 Emotion Detection Using CNN (Deep Learning Project) | Built a CNN-based facial emotion recognition system in TensorFlow/Keras achieving 87% accuracy across 7 emotions. Applied OpenCV preprocessing and data augmentation, improving model generalization and reducing overfitting by 25%. | TensorFlow, Keras, CNN, OpenCV, Deep Learning |
Want more highlights? 📌
- 🧪 Sparkify Labs: EMR + PySpark Streaming, ALS Recommender, MapReduce jobs
- 🛰️ Space Research DBMS: SQL schema design, ER modeling, query optimization
- 🧭 Virtual Assistant (Android): Accessibility features, real-time location & weather
- 👁️ Emotion Detection (CNN): End-to-end pipeline with model serving (extended work)
- 📦 Production-grade ETL with Spark + Airflow (MWAA)
- 🔍 RAG-powered real-estate data analytics
- 🧠 LLM-driven review authenticity scoring (Uplaud AI)
- 🌐 Interactive dashboards with React + AWS serverless
