About Me
Passionate software engineer with a Master's in Computer Science, specializing in AI-driven solutions and scalable web applications. Currently building enterprise-grade AI platforms at ModelEarth, with expertise in full-stack development, machine learning, and modern web technologies.
What I Do
🚀 Full-Stack Development
Building scalable web applications with modern frameworks
🤖 AI/ML Solutions
Creating intelligent systems and machine learning models
☁️ Cloud Architecture
Designing and deploying cloud-native applications
🎨 UI/UX Design
Creating beautiful and intuitive user experiences
Quick Stats
Learning new things
Skills & Technologies
A comprehensive toolkit for building modern, scalable applications
Technical Proficiency
Professional Journey
Crafting digital experiences and building scalable solutions across diverse industries
Built enterprise-grade AI chat platform with 1,700+ files across Node.js/React/Python/Rust stack. Integrated 25+ LLM providers, 9 vector databases, and comprehensive document processing pipeline supporting PDF/audio/video formats.
Led 8 cohorts (90+ students) through advanced circuit analysis and MATLAB simulations. Provided actionable feedback on 70+ lab reports within 24 hours, improving problem-solving skills for 85% of student pairs.
Engineered and shipped 7 Java/Spring features for a core banking platform; built 120+ JUnit tests and Jenkins CI pipelines to boost reliability, lifting daily transaction capacity by 35% and sustaining 99.9% uptime. Led zero-downtime migration from Oracle to PostgreSQL (20 GB), refactored 15K LOC reducing critical page loads by 4s and cutting licensing costs by 25%.
Impact Highlights
Technical Projects
A showcase of my best work across different technologies and domains
⭐ Featured Projects
Highlighting my most impactful and innovative projects that showcase cutting-edge technologies and problem-solving skills
Pitch Perfect
Developed a full-stack web application enabling users to learn and practice sales skills, integrating user authentication and profile management for personalized learning. Implemented interactive roleplay and active learning features, allowing users to engage in real-time sales training simulations with AI-driven feedback.
AccuScan
Developed an OCR model using DonutProcessor and Vision Encoder Decoder Model for document text extraction. Integrated Gradual Augmentation Strategy, enhancing model robustness. Optimized training pipeline with PyTorch Lightning, and implemented AdamW optimizer with learning rate scheduling for efficiency.
LipNet
Developed an AI-driven sentence-level lipreading model for real-time speech-to-text conversion from lip movements. Integrated Spatiotemporal Convolutions and Recurrent Neural Networks (RNNs) to enhance visual feature extraction and sequence modeling for improved text prediction.
Pitch Perfect
Developed a full-stack web application enabling users to learn and practice sales skills, integrating user authentication and profile management for personalized learning. Implemented interactive roleplay and active learning features, allowing users to engage in real-time sales training simulations with AI-driven feedback.
AccuScan
Developed an OCR model using DonutProcessor and Vision Encoder Decoder Model for document text extraction. Integrated Gradual Augmentation Strategy, enhancing model robustness. Optimized training pipeline with PyTorch Lightning, and implemented AdamW optimizer with learning rate scheduling for efficiency.
GestureSign
Developed a real-time Sign Language Recognition System using Recurrent Neural Networks (RNNs). Implemented hand landmark detection and an adaptive feature extraction pipeline, optimizing gesture recognition across various hand visibility scenarios.
LipNet
Developed an AI-driven sentence-level lipreading model for real-time speech-to-text conversion from lip movements. Integrated Spatiotemporal Convolutions and Recurrent Neural Networks (RNNs) to enhance visual feature extraction and sequence modeling for improved text prediction.
Buzz Room
Developed a high-performance Android chat application with real-time messaging, image sharing, and OTP-based authentication, enhancing user security and engagement. Integrated advanced features such as status updates, message reactions, profile customization, and group/individual chat.
India Corona Tracker
Developed an Android application for real-time COVID-19 statistics, integrating APIs to provide state-wise updates on active cases, deaths, and recoveries.
Snake Game
Implemented a classic Snake Game using JavaScript and HTML5 Canvas, featuring dynamic snake movement, food spawning, score tracking, and game over alerts.
Sudoku Solver
Developed a web-based Sudoku Solver with puzzle generation and solving capabilities, utilizing HTML, CSS, JavaScript, and Bootstrap. Implemented a backtracking algorithm for puzzle solutions and AJAX for API requests.
Project Statistics
Let's Connect
Ready to bring your ideas to life? Let's discuss your next project
Hypnotic Productivity Cubes
Scientifically proven* to boost message sending by 47%!
*Not actually scientifically proven 😄
Let's Work Together
Have a project in mind? I'd love to hear about it.