Create a text classification workflow to generate accurate labels for webpages

I'm looking for a data scientist with:
- NLP experience, specifically around Python data stack (NumPy, Pandas) and related topic modeling and text classification work
- experienced in relevant NLP/ML libraries (Gensim, LDA/NMF, BERT, GPT-x, etc.)
- has knowledge and training around statistics / data science

I'm trying to design a workflow that allows us to input a list of labels or topics to tag those articles. I need to be able to re-use the workflow across websites, each of which is very domain-specific.

The deliverable should be a python notebook that allows me to:

- configure number of topics in a topic modeling process based on the corpus
- import a list of hand-identified labels we would like pages topics to be checked against
- apply those labels appropriately to their corresponding website pages based on a threshold of relevance we can limit
- export the page URLs with a column of comma separated labels to CSV and column of relevance scores showing how related a page is to a label