Education
Kalinga Institute of Industrial Technology (KIIT), B.Tech in Electronics and Computer Science | CGPA-8.57
2022 – 2026 | Bhubaneswar, Odisha
Internship Experience
Mar 2024 – Nov 2024 | Remote
  • •Developed an Android app for multicast audio streaming, supporting 1 host and up to 5 simultaneous client devices over Wi-Fi, using NSD for dynamic device discovery and selective connections.
  • •Implemented TCP and UDP-based audio streaming, achieving synchronized playback within 100ms across devices and real-time microphone streaming with <200ms latency.
  • •Built reliable client-host connections using socket programming, handling connection requests, acceptances, and error recovery for over 50 successful test sessions.
  • •Optimized network throughput by 70%, increasing transmission rates from 125–140 kbps to 200–230 kbps via dynamic packet size tuning and link speed adjustments, verified through Wireshark analysis.
  • •Collaborated on a Python + SQLite backend, hosted on AWS EC2, to manage device sessions and audio metadata, supported by a CI/CD pipeline for automated testing, debugging, and deployment.
    • Projects
    MultiDoc-AI-Assistant⁠, (Python, Streamlit, Langchain, Groq API, RAG, HuggingFace's BAAI/bge-small-en-v1.5))
  • •Engineered a GenAI application for advanced conversational search and interactive querying across diverse, multi-modal sources like PDFs, websites, and handwritten notes/images.
  • •Implemented robust backend data pipelines for data extraction using PyPDF2, BeautifulSoup, and Google Cloud Vision OCR, optimizing resource use with session-based rate limiting.
  • •Architected a Retrieval Augmented Generation (RAG) system with embeddings and semantic search using a ChromaDB vector database, significantly enhancing information retrieval accuracy.
    • LLM-Powered Dynamic Test Platform⁠, (Flask, Python, Firebase, LangChain, LangGraph, JavaScript, Chart.js, Render)
  • •Engineered a full-stack, AI-powered web application to generate dynamic aptitude (MCQ) and programming (DSA/​SQL) questions, leveraging the Groq API with Llama 3.1 to provide a limitless, personalized practice environment for students.
  • •Implemented an agentic AI system using LangGraph and LangChain to analyze student MCQ performance, autonomously generating detailed feedback reports with curated links to external learning resources, significantly enhancing the learning feedback loop.
  • •Developed an AI-powered code evaluation feature where users submit solutions in Python, C++, or Java within an interactive editor; the backend AI then provides expert-level feedback on correctness.
  • •Architected a secure, role-based user management system (Student/Teacher) utilizing Firebase Authentication and Cloud Firestore to store user data and quiz histories, enabling a teacher dashboard with student analytics and data visualization.
    • Skills
    Programming Languages: C/C++, Python, Java
    Frameworks: FastAPI, Flask, Android SDK
    Relational/NoSQL Databases: MySQL, Redis
    Version Control & Tools: Git, GitHub, Android Studio, Postman, Docker, Linux
    AI/ML & LLMOps: NLP, LangChain, LangGraph, LangSmith
    Coursework Subjects: Computer Networks | Database Management Systems | Object-Oriented Programming (OOPs) | Operating Systems | Data Structures & Algorithms (DSA)
    Achievements
  • •Selected for Keploy API Fellowship (top 1,000 of 18,500 applicants), gaining hands-on experience in open-source contributions & software testing.