I started tinkering with code at 13, and have been obsessed with creating things on the internet ever since.
My journey took me through different eras, starting around 2016 when React had just begun. Along the way I learned that the best products come from understanding users and a relentless focus on quality.
When I'm not coding, you'll find me contributing to open source, writing about tech, or exploring new technologies. Big believer in building in public.
Experience ⚒
Research Intern
National Institute of Technology · Silchar
Worked under Dr. Ram Kumar Karsh on a multimodal deep learning pipeline that predicts coronary artery disease from OCTA retinal imaging and clinical data, fusing self-supervised (SimCLR) ConvNeXt encoders, segmentation-derived vascular biomarkers, and a tabular MLP into a calibrated eight-cohort classifier, and benchmarking six segmentation architectures (U-Net, U-Net++, Attention U-Net, DeepLabV3+, SegFormer) within the same controlled pipeline.
Google Developer Group on Campus - Guahati University · Guwahati
Leading the web development team to create and maintain web applications for various campus events and initiatives.
Full Stack Developer Intern
FRINT · Guwahati
Built an enterprise-grade AI proctoring system as a multi-service monorepo: React + Vite frontend, Express/Socket.io gateway, Python FastAPI microservices for AI/ML processing (Whisper STT, MediaPipe/OpenCV for computer vision), MongoDB + Redis for data, and Groq and Gemini for intelligent candidate monitoring and analysis.
Worked under Dr. Ram Kumar Karsh to develop a real-time Indian Sign Language recognition system using a hybrid 2D Convolutional Neural Network + LSTM architecture trained on a custom video dataset.