Resume parsing + Cv analysis

Follow the full discussion on Reddit.
Hi ! so for my final year project I will be working on a cv parser and matching cvs with job postings, I'm thinking about fine tuning LayoutLM on my cvs dataset( of 5000 resumes or so not yet labeled) to get the structure of a resume (contact info , skills , education , etc) and then combine it with NER to identify the details in each section (name , uni name , date of start etc ) . Is it good enough or should I take another approach ? Or how would you tackle the problem ? feel free to share any ideas u have about this project Thank you !