Experience
Teaching Assistant
IIT Indore
August 2025 – Present • Education / Communication Systems
- TA under Prof. Dibbendu Roy for the course EE319 - Design and Analysis for Communication Systems
- Assisted in designing and grading challenging quizzes
- Conducted tutorials on topics including Random Variables, Random Processes, and Queuing Theory
Tech Stack:
Teaching Communication Systems Probability Theory
Developer Intern
Samsung R&D Institute India
May 2025 – July 2025 • 6G Communications / Machine Learning
- Collaborated with the 6G Standards Team to develop novel machine learning–based methods for CSI-RS (Channel State Information Reference Signal) processing
- Designed and implemented UNet-inspired deep learning architectures, achieving a Normalized Mean Squared Error (NMSE) in the order of 1e-3
- Work under patent filing process for its potential impact on future 6G physical layer standards
- Received return offer from Samsung for the Advanced Developer Role
Tech Stack:
Python PyTorch Deep Learning 6G Communications Signal Processing
Research Intern
Prof. M. Tanveer | OPTIMAL Research Group, IIT Indore
May 2024 – Present • Deep Learning / Audio-Visual Speech Enhancement
- Developed LSTMSE-Net, an audio-visual speech enhancement model to isolate and enhance speaker audio in noisy environments
- Engineered a temporal feature extraction pipeline using RNN and LSTM units to jointly model audio-visual dependencies
- Achieved a 3× reduction in inference time compared to the baseline model with improvements in speech quality
- Original paper accepted in InterspeechW 2024. Currently working on an advanced version using ConvNeXtV2 based video pipeline and an audio decoder inspired from deep state space modelling
Tech Stack:
Python PyTorch Deep Learning LSTM RNN Audio Processing Video Processing
Research Intern
Prof. Nagendra Kumar | LIPG, IIT Indore
May 2023 – Present • Deep Learning / Medical Image Segmentation
- Developed novel U-Net based deep learning models for liver tumor segmentation, handling data pre-processing and creating custom callbacks, metrics, and loss functions
- Implemented advanced architectures like Squeeze-and-Excitation Networks and Atrous Spatial Pyramid Pooling, achieving a state-of-the-art accuracy of 98%
- Co-authored a research paper detailing the methods and results. Published in Elsevier Biomedical Signal Processing and Control
Tech Stack:
Python PyTorch TensorFlow Deep Learning U-Net Medical Imaging Computer Vision
Research Contributor
Dr. Debesh Jha | Northwestern University
May 2024 – July 2024 • Deep Learning / Medical Image Segmentation
- Implemented various liver tumor segmentation models including DeepLabv3+, UNet, and HiFormer-L on the LiTS dataset using PyTorch and TensorFlow
- Engineered custom, modular PyTorch data loaders and transformation pipelines for the LITS dataset
- Source code available at GitHub
Tech Stack:
Python PyTorch TensorFlow Deep Learning Medical Imaging Computer Vision