AI/ML · Data Science · Full-Stack
B.Tech CSE @ KIIT Bhubaneswar · CGPA 8.40 · Batch 2027
I build AI-powered systems — from RAG pipelines and computer vision models to full-stack web apps. Currently in my placement year, targeting AI/ML, Data Science, and SWE roles. I care about shipping things that actually work, not just things that look good on paper.
- 🏆 Top 50 — Smart India Hackathon 2025
- 💻 400+ DSA problems solved (LeetCode + GFG)
- 🎓 Google Data Analytics Professional Certificate · HackerRank SQL · Cisco NetAcad
- 🎓 Data Science — CodeWithHarry · Power BI — ExcelR
TRAVELMAiT V2 — AI Travel Planner
Top 50 Smart India Hackathon 2025 · Live: travelmait-v2.vercel.app
Mood-based itinerary generation for offbeat Odisha destinations, group travel matchmaking, and multi-modal transport sorting. Built on a curated dataset of 100 Odisha locations with RAG-powered recommendations.
React Tailwind Flask MongoDB Atlas Groq (LLaMA-3.3-70b) Foursquare API JWT
StudyMind — RAG Teaching Assistant
Upload PDFs and ask questions in natural language. Full RAG pipeline with semantic chunking, HuggingFace embeddings (all-MiniLM-L6-v2), and Groq-powered LLaMA responses. Built end-to-end in 2 days.
FastAPI PostgreSQL React Tailwind Groq (LLaMA) HuggingFace Python
Optic Disc Detection — Retinal Image Segmentation
Comparative study of 4 methodologies for optic disc localization in fundus images — U-Net CNN, K-Means, DBSCAN, and traditional morphological CV. Best result: 92.92% Dice / 87.97% IoU with U-Net.
| Method | Dice | IoU | Pixel Accuracy |
|---|---|---|---|
| U-Net CNN | 92.92% | 87.97% | 99.80% |
| K-Means | 85.41% | 77.98% | 99.34% |
| DBSCAN | 84.34% | 76.40% | 99.25% |
| Traditional CV | 83.71% | 76.83% | 98.76% |
PyTorch TensorFlow OpenCV Scikit-learn U-Net DBSCAN K-Means
Customer Feedback Sentiment Analysis — NLP + BI
End-to-end sentiment analysis pipeline on customer reviews — preprocessing, feature engineering, classification model, and an interactive Power BI dashboard surfacing actionable business insights.
Python Pandas NLTK Scikit-learn SQL Power BI
DPI Engine — Deep Packet Inspection in Java
A from-scratch deep packet inspection engine covering pcap parsing, TLS SNI extraction, DNS query analysis, and TCP connection tracking. Built as a serious networking systems project for placement prep.
Java Networking TLS pcap TCP DNS SNI
Languages · Python · Java · JavaScript · SQL · C
AI / ML / Data · PyTorch · TensorFlow · Scikit-learn · Pandas · NumPy · OpenCV · HuggingFace · LangChain · NLTK · Power BI · Tableau
Web & Backend · React · FastAPI · Flask · Node.js · Tailwind CSS
Databases · PostgreSQL · MongoDB · MySQL
Tools · Git · GitHub · Docker · Jupyter · Google Colab · VS Code · Linux
Open to AI/ML, Data Science, and SWE internship opportunities