kailash N

I am a Computer Vision Engineer with two years of hands-on experience building AI-driven systems for object detection, OCR, and facial recognition. I design custom deep learning models, implement end-to-end computer vision pipelines, and work with real-world datasets to deliver production-ready solutions.

My projects include a PyTorch-trained YOLO model for shipping port monitoring with centroid-based tracking and OCR for container IDs, an office monitoring system for counting and PPE compliance using pose estimation and object detection, and an OCR-based identity document and invoice verification system deployed with Streamlit. I also built a CNN driver drowsiness detector and several ML projects such as petrol price regression and Twitter sentiment analysis.

I bring a strong foundation in Python, machine learning, and computer vision frameworks (OpenCV, TensorFlow, PyTorch), along with practical experience using Docker, SQL, NumPy, Pandas, and deployment tools. I focus on reliable, interpretable solutions that solve operational problems in real environments.