0 km/h
ARNAB DAS
Contact

Code &
VISION

The Specialist

Engineering
Intelligence

I am Arnab Das. A dual-degree student constructing the bridge between Data Science and Engineering. I specialize in extracting signals from noise, whether it's detecting rare Z-Prime boson anomalies or architecting real-time communication platforms.

Analyzing particle collisions at IIT Patna. Building scalable platforms at IIT Madras. My work oscillates between the precision of scientific computing and the creativity of full-stack engineering. I engineer intelligent systems that solve complex problems with both mathematical rigor and intuitive user experience.

GPA 8.98
Python / JS / Java
The Arsenal

Technical
Proficiency

Languages

Python, C++, Java, JavaScript, HTML, CSS, SQL

Frameworks & Libraries

TensorFlow, PyTorch, Scikit-Learn, NumPy, Pandas, Matplotlib, React.js, Node.js, Express, Socket.IO

Machine Learning & AI

Autoencoders, GBM (XGBoost/LightGBM), SVM, Neural Networks, Computer Vision, Anomaly Detection

Tools & DevOps

Git, GitHub, Docker, MongoDB, MySQL, Postman, Vercel

Projects
01

Z-Prime

Anomaly Detection
Designed a Deep Autoencoder architecture in TensorFlow/Keras to model latent representations of Standard Model background data. Achieved 50% Signal Efficiency with a background rejection factor of ~9.2 in unsupervised environments, enabling the detection of unseen particle physics anomalies without predefined labels.
VIEW
02

Collision Hunt

ML Pipeline
Engineered a high-performance ML pipeline using Uproot and Awkward Array to process vectorized Monte Carlo collision data. Constructed Lorentz-invariant kinematic features and deployed XGBoost classifiers to maximize discrimination of rare signal events, validated via statistical "Bump Hunt" analysis.
VIEW
03

LynqIt

Real-time Platform
Built a robust real-time messaging ecosystem using the MERN Stack and Socket.IO, featuring AI-powered auto-replies and secure JWT authentication. Optimized for low-latency message delivery, successfully handling 100+ concurrent users in a scalable, research-backed implementation.
VIEW
Next Steps Start a
Conversation
GITHUB LINKEDIN
Built and Designed by Arnab Das