Ashish Sharma

Designed a CNN-based acoustic recognition system classifying 8 ship types at 91% accuracy and estimating distance with 3% MAE using spectrograms and VMD-enhanced multi-domain features for real-time sonar processing.

Built a real-time fog-aware AI vision system using CLAHE and adaptive preprocessing with 89.7% accuracy and GPU-accelerated SVM classification reducing latency to 40ms for edge deployment.

Developed a hybrid ML pipeline (XGBoost, LightGBM) for wind forecasting that reduced prediction error by 18% and improved smart-grid scheduling precision by 22%, deployed for real-time inference on Google Cloud.

Grade: CGPA - 7.34

Activities and societies: Men’s Hockey Team member, NIT Rourkela (2023–Present)

B.Tech. in Electronics and Instrumentation Engineering ongoing at NIT Rourkela with coursework in deep learning, computer vision, and cloud computing; current CGPA 7.34.

Grade: 94.20%

Completed AISSCE (RBSE) with focus on Science (Physics, Chemistry, Mathematics) achieving 94.20% in May 2021.

Grade: 85.67%

Completed AISSE (RBSE) in May 2019 with an overall percentage of 85.67%.