AI-powered CV platform, which helps to match candidates with companies based on NLP models, which enable us to automatically parse candidate’s resume, extract contact details and skills. Deliverables:
- Name, address, addresses extractors based on Stanford core NLP, NLTK tools.
- Skills parser based on custom skill-set model, that produced normalized-form skills and word2vec vectors to match resumes on the proximity of skills, match proposals from companies with candidate CV based on skills.
- Highload service to manage relearning and work of ML model, including parallelism, caching, graylog server to monitor performance
Technology stack:
Python 3, Flask, Celery, DynamoDB, Redis, Tensorflow, Keras, NLTK, log4j, graylog, swagger, AWS Lambda, AWS triggers.