SANTHOSH KUMAR M

I’m a B.Tech student in Artificial Intelligence and Data Science with hands-on experience in deep learning, computer vision, and AI model development. I trained at ISRO Propulsion Complex (IPRC) on real-world AI applications for aerospace, building solutions that move from detection to actionable insights.

At IPRC, I designed an AI-based system to analyze rocket engine exhaust flame geometry using video data. I implemented YOLO-based deep learning models for real-time flame detection and segmentation, accelerated training/inference with PyTorch on NVIDIA RTX GPUs using CUDA Toolkit, and extracted combustion parameters like flame length, area, combustion efficiency, and exhaust behavior.

I also built a non-contact diagnostic pipeline using Python, OpenCV, and Streamlit for live visualization. Beyond my trainee work, I’ve delivered projects such as brain tumor detection from MRI scans using VGG16 transfer learning with Grad-CAM explainability (deployed via Streamlit), and AI analytics and agricultural computer vision solutions, earning recognition including a TNSCST Rs. 7,500 grant and hackathon/presentation prizes.