This study provides a novel resume parsing solution using a hybrid Spacy Transformer BERT and Spacy NLP methodology. The main goal is to create a resume parser that can efficiently extract pertinent data from unstructured resumes that do not adhere to a predetermined resume structure and may contain information presented in a non-standardized manner. We also intend to investigate the usage of video resumes as a fresh source of candidate data and put forth a cutting-edge method for video resume parsing that combines visual and audio processing methods. We employed a hybrid methodology of Spacy Transformer BERT and Spacy NLP to accomplish these goals. A pre-trained deep learning model called Spacy Transformer BERT captures the text's semantic meaning, and Spacy NLP employs natural language processing to glean pertinent information from it. Our method combines the strengths of the two models for high accuracy and efficiency in collecting pertinent information from resumes. Using a dataset of resumes, we ran experiments to gauge how well our suggested system performed. The outcomes demonstrate that our system was highly accurate in retrieving pertinent data, including candidate names, contact information, qualifications, work experience, and other pertinent characteristics.