FULL-STACK DEVELOPER / SYSTEMS ENGINEER
Full-Stack Developer with a passion for building scalable and production-grade applications. I have a strong background in backend development, system design, and cloud-native technologies.
Currently pursuing Dual Degree (B.Tech & M.Tech) at IIT Delhi. Experienced in building distributed systems, real-time applications, and ML-powered tools. I solve problems with minimal dependencies and maximum efficiency.
"Build systems that scale, code that's maintainable, and solutions that matter."
{
"languages": [
"Python", "C++", "JavaScript",
"TypeScript", "Go", "Java"
],
"frontend": [
"React", "Next.js", "TailwindCSS",
"CodeMirror", "Recoil"
],
"backend": [
"Node.js", "Express.js", "Flask",
"Socket.IO", "gRPC", "REST APIs"
],
"ml_data": [
"PyTorch", "TensorFlow", "Scikit-learn",
"Pandas", "NumPy", "spaCy"
],
"databases": [
"PostgreSQL", "MongoDB", "MySQL",
"Redis", "RocksDB"
],
"devops": [
"Docker", "Kubernetes", "AWS",
"Nginx", "Prometheus", "Grafana"
],
"tools": [
"Git", "Linux", "CI/CD"
]
}
Production-grade HTTP load balancer with multiple balancing algorithms (Round Robin, Weighted, Least Connections), sticky sessions, path-based routing, and real-time monitoring dashboard using WebSockets.
Node.js, Express.js, Socket.IO, Winston
Real-time collaborative code editor enabling multiple users to code together. Features live cursor tracking, room-based collaboration, file management, and syntax highlighting for multiple languages.
React, Node.js, Socket.IO, CodeMirror, Recoil
Intelligent autoscaling system reducing infrastructure costs by 35%. Uses predictive scaling algorithms with time-series analysis to anticipate traffic spikes, achieving 60% faster scale-up response.
Python, Kubernetes, Prometheus, Grafana
AI-powered tool to parse and highlight PDF resumes using weighted ensemble of transformer models (spaCy, MiniLM). Features three-tier color-coded highlighting and a QA chatbot for resume questions.
Python, Flask, PyTorch, spaCy, Transformers
Deep learning image captioning system using VGG16/MobileNetV2 encoder with LSTM-based decoder and attention mechanism. Trained on Flickr8k dataset with Streamlit deployment interface.
Python, TensorFlow, Streamlit, NLTK
Full-stack AI application that converts natural language to executable SQL using Gemini/OpenAI. Features schema-aware generation, interactive database explorer, and real-time query execution with result visualization.
Python, FastAPI, Streamlit, SQLAlchemy, MySQLAxtria - Ingenious Insights | May 2025 - July 2025
FidelFolio Investment Pvt Ltd | May 2024 - July 2024
IIT Delhi | 2021 - 2026
$ echo "Open to full-time opportunities in full-stack, backend, and distributed systems."