HELLO FOLKS!

Developed a peer-to-peer file sharing application that transfers complete files directly between peers. It uses Merkle tree for correctness verification and optimisation Implemented automatic peer discovery on local networks, file search across connected peers, and a simple command-line interface for easy interaction. Managed direct TCP connections for reliable file transfers and ensured file integrity with minimal overhead. Built entirely with Python standard libraries, enabling lightweight and dependency-free deployment,used flask based frontend to add or see online peers

A Chrome extension to enhance browsing experience by customizing website appearance. Features include dark mode, font and style adjustments, brightness control, and high-contrast mode. Implemented a responsive popup UI for instant changes, with offline functionality and user preference storage for seamless personalization.

Led a comparative study evaluating 27+ ML model–feature combinations for suicide-ideation detection using a dataset of 13,000+ social-media posts. Implemented TF-IDF, BoW, CHI2-1000, and SVD-300 embeddings, and ran PCA/t-SNE experiments to measure the impact of dimensionality reduction. Developed geolocation and NER pipelines to map crisis locations across platforms. Built similarity matrices and DGIM/Sliding-Window stream algorithms for Instagram user-risk analysis. Achieved the best performance with TF-IDF + Logistic Regression, outperforming boosting models by 5–12%.

Developed a full-stack Leave Management System handling student leave requests and admin approvals. Implemented versions using MERN stack (MongoDB, Express.js, Node.js, React, JWT) and C#/.NET with Azure Web Apps, plus a lightweight HTML/CSS/JS demo. Integrated frontend with Azure Function-based backend APIs, managed CORS, and deployed serverless functions for scalable, secure, and low-maintenance production. .

Developed a series of projects on the ToyRISC ISA, including writing assembly programs, building an assembler, and implementing single-cycle, pipelined, and cache-integrated processor simulators in Java. Enhanced the simulator with discrete-event modeling, hazard handling, and performance analysis (IPC, stalls, cache impact), gaining hands-on experience with computer architecture design and evaluation.

Built an end-to-end compilation workflow producing optimized low-level assembly code. Implemented all six compiler phases: Lexical Analysis (tokenization), Syntax Analysis (AST generation), Semantic Analysis (type & variable checks), Intermediate Code Generation (three-address code, quadruples, triples), Code Optimization, and RISC Code Generation using Python 3 and object-oriented design.

Developed high-performance matrix multiplication with BLAS and SSE/SSE3 intrinsics, optimized array summation to minimize cache misses, and debugged matrix operations using GDB and Valgrind. Designed modular grid-based PDE solvers and implemented generic C++17 containers with memory-safe practices, operator overloading, and vectorized computation for scientific applications.

Developed an interactive AI assistant on Google Colab capable of web search, hotel booking predictions, student placement & salary predictions, and weather forecasting. Applied machine learning, NLP, web scraping, and data preprocessing to deliver accurate, real-time predictions and responses.

Implemented a Python SMTP client using raw sockets and TLS to authenticate and send emails programmatically. Demonstrates understanding of TCP socket communication, SMTP protocol commands, and secure authentication using base64 encoding.

Developed a Python TCP server that efficiently receives files from clients, extracts the first and last 10 lines using collections.deque, and sends the processed content back. Optimized for memory-efficient data handling and reliable network communication.

Developed a turn-based Checkers game with real-time synchronization over TCP sockets. Implemented game state management, move validation, and client-server communication to enable interactive gameplay between two players.

Built a suite of C/C++ systems programming simulators covering image processing, process management, CPU scheduling, parallel pipelines, and virtual memory management. Implemented multi-threaded and multi-process designs with shared memory, pipes, semaphores, and atomic operations, while measuring performance across various workloads.