Artificial intelligence - Building a Software with Neural Network and Machine Learning

The main utility of the software will be to analyze the professional profile for each client and to search for appropriate internet vacancies for each profile. Machine Learning | Cosine Similarity | Neural Network | Web Scrap | Neural network model takes care of every single text written in document. Neural network model will be reliable and robust. First step: The program will review the Customer Professional History Questionnaire. (These documents will have the client's professional background, skills and professional activities, professional achievement, software skills, and educational background.) Second step: Will access some websites to find jobs - that are associated with a company and not confidential jobs - that fit with the professional selected customer profile. (That fit at least 90% with the professional profile of the selected client). (Some of these sites require Login and Password that I have). Third step: Report the URL of each vacancy found that fits into the profile of the selected customer. (Do not report URLs of vacancies already reported to that client, if possible, do not report vacancies already expired). Fourth step: An internal search engine. Where I would write the description of a vacancy and the system would tell me which client has the most appropriate professional profile for the vacancy.