Dávid Kocsis

Innovative and results-driven Software Engineer with 3+ years of experience in embedded systems development, algorithm optimization, and automation within high-end automotive projects. Strong problem-solving skills, rapid learning ability, and solid expertise in Python, C++, and computer vision. Successfully designed and implemented high-performance simulation environments, introduced research-based algorithms, and significantly reduced debugging time through automated solutions. Open to opportunities in Artificial Intelligence and data-driven technologies, combining analytical thinking with hands-on development experience.

• Automatic Number Plate Recognition (ANPR) – BSc Thesis
  Implemented a TensorFlow-based neural network for license plate detection and OCR-based character recognition. Prepared and augmented datasets, trained the model, and integrated OpenCV image processing algorithms to improve accuracy.
• Asphalt and Lane Detection with Semantic Segmentation – MSc Thesis
  Developed a PyTorch-based deep learning model for asphalt surface detection, achieving high accuracy even on low-quality datasets. Applied computer vision algorithms (NumPy, OpenCV) for lane detection on the segmented surfaces. Gained extensive experience in neural network development, image processing, and model evaluation.