I'm a CS student who loves building real systems — from federated identity architectures and AI-powered document search to data analytics engines and ML predictors. I work across the full stack with Python, .NET, Node.js, and modern AI tooling.
- 🔭 Currently building federated SSO systems and RAG pipelines
- 🔐 Architected a modern redirect-based federated SSO microservice system (IdP & SP pattern)
- 🤖 Built a retrieval-augmented generation app with ChromaDB & Groq
- 📊 Experienced in data analytics, visualization, and ML prediction
- 📚 Avid reader who loves diving into new topics
- 🎮 Gamer in my downtime | ⚽ Sports enthusiast
- 💼 Open to collaboration and opportunities
- 🧠 Interested in: Security & Identity, AI / RAG Systems, Data Science, Web Development
| 🔐 Redirect-Based Federated SSO | 🤖 AI RAG System |
|---|---|
| SSO A modern Redirect-Based Federated Single Sign-On (SSO) Architecture built for distributed microservice environments. Two decoupled microservices — App 1 (Identity Provider / IdP) running on Node.js/Express + MongoDB, and App 2 (Service Provider / SP) running on Express — implement centralized OAuth-style login redirections with secure JWT session validation. 🔹 Centralized IdP login redirect with dynamic returnUrl routing🔹 Loop-free account switching & session purging via clearCache=true🔹 HMAC SHA-256 signed JWT verification across decentralized microservices 🔹 Interactive Architecture Portal with automated PowerShell startup       |
AI_RAG A retrieval-augmented generation app built around a local ChromaDB vector store, PDF ingestion, sentence-transformer embeddings, and Groq for answer generation. Load any PDF, index it into vector chunks, and ask natural language questions — the system retrieves the most relevant context and generates accurate answers via LLM. 🔹 PDF text extraction & overlapping chunk splitting 🔹 Sentence-transformer embeddings stored in ChromaDB 🔹 Cosine similarity retrieval for relevant context 🔹 Groq LLM for final answer generation     |
| 📊 GDP-Analysis_Phase2 | 📈 Performance-Marketing-ROI-Predictor |
|---|---|
| Phase 2 SDA Project — a Python-based data transformation and analytics engine for global GDP trends. Features data ingestion from CSV/JSON, ISO-based cleaning, and 7+ analytical tasks: top/bottom GDP rankings, growth rates, continent averages, global trend analysis, fastest growing continents, consistent decline detection, and GDP contribution breakdowns. Supports console and chart output via a plugin system. | ML-based performance marketing ROI prediction using Jupyter Notebooks. Analyzes marketing campaign data to predict return on investment, helping make data-driven budget allocation decisions. |
   |
 |
👉 See pinned repos below for live demos and code!
| 🔐 Security & Identity | 🤖 AI & RAG | 📊 Data Analytics | 📈 ML & Prediction |
|---|---|---|---|
| Architected a modern redirect-based federated SSO architecture with centralized IdP/SP routing and loop-free account switching | Built a full RAG pipeline — PDF to embeddings to LLM-powered answers | Engineered a GDP analytics engine with 7+ analytical modules & plugin-based output | Developed a marketing ROI predictor with Jupyter & ML |
Languages
Frameworks & Libraries
AI & Data Science
Databases & Backend Services
Developer Tools
"Code is like humor. When you have to explain it, it's bad."