Mansi Ambwani

I’m a B.Tech Electrical Engineering student at the Indian Institute of Technology, Bhilai, focused on using machine learning and deep learning to solve real signal and data problems. I learn fast, iterate quickly, and take ownership from data collection through evaluation.

As a Research Intern, I built a Python-based hardware SDR pipeline and automation scripts across multiple modulation schemes, generating and labeling large datasets for training and evaluation. I also developed a CNN-based Automatic Modulation Classifier using higher-order cumulants (HOC) and a ResNet architecture, reaching 96–99% classification accuracy, and built an end-to-end pipeline for channel estimation for RIS using USRPs.

My projects range from ML forecasting and classification (including time-series demand forecasting with Prophet) to engineering-focused simulation work like an EV bidirectional DC-DC converter and a 3-bit Flash ADC design. I’m excited to bring this end-to-end research mindset to a role where I can collaborate and ship reliable solutions.