π B.Tech in Computer Science & Engineering (2025) Β |Β π» Full Stack Development Β |Β π€ AI β’ NLP β’ LLMs β’ Computer Vision
- π Building AI-powered applications focused on OCR, document intelligence, NLP, and Large Language Models.
- π» Passionate about Full Stack Development, developing scalable web applications using modern frontend and backend technologies.
- π§ Interested in Artificial Intelligence, Machine Learning, Natural Language Processing, and Intelligent Document Processing.
- π I enjoy learning new technologies by building real-world projects from scratch and understanding how modern AI systems work internally.
- π― Currently preparing for graduate research in Artificial Intelligence and Intelligent Computing while continuously improving my software engineering skills.
- π€ Always open to collaborating on AI, Full Stack, and open-source projects.
| Project | Description |
|---|---|
| π Urdu Heritage Translation Using OCR β Final Year Research Project | Research project titled "Reviving Historical Documents for Modern Understanding." Developed an OCR-based workflow for digitizing historical Urdu manuscripts using image preprocessing, OCR, and Urdu-to-English translation. The repository includes the research paper, project report, presentation, system architecture, workflow, and comprehensive documentation for preserving historical Urdu documents. |
| π Urdu Document Assistant | Local full-stack document processing system built with FastAPI, Next.js, PostgreSQL, Docker, and Tesseract OCR. Supports secure authentication, document upload, OCR-based Urdu text extraction, Urdu-to-English translation, and document management for scanned historical documents. |
| π AI Resume Analyzer | AI-powered resume analysis tool that evaluates ATS compatibility, extracts resume content, identifies missing skills, and generates personalized improvement suggestions using NLP techniques and LLM APIs. |
| π§© Mini LLM From Scratch | Educational implementation of a GPT-style Transformer language model in PyTorch covering tokenization, embeddings, positional encoding, self-attention, training, and text generation to understand Large Language Model fundamentals. |
| π¬ Socket Chat Application | Real-time client-server chat application built using socket programming and WebSockets, enabling instant communication between multiple users through a simple and responsive interface. |
| βοΈ AI Debate Akhara | Interactive AI-powered debate platform where two AI personalities discuss user-selected topics using prompt engineering and LLM-based conversations within an engaging web interface. |
β Thanks for visiting my profile! I'm always excited to learn, build, and collaborate on meaningful software and AI projects.