NLP Engineer for Intelligent Resume Screening System

We are looking for an NLP Engineer to develop a resume matching engine that scores candidates against job descriptions with high semantic accuracy. Unlike simple keyword matchers, this system must understand context (e.g., "React" vs "React Native") and generate explainable scoring reports. The project involves parsing unstructured CVs, extracting entities, calculating semantic similarity embeddings, and serving the results via a dashboard-ready API. Key Responsibilities - Resume Parsing: Implement a robust pipeline to convert diverse resume formats into structured JSON schemas (Skills, Experience, Education) - Embedding Logic: Use Hugging Face sentence transformers to generate vector embeddings for both resumes and job descriptions - Scoring Engine: Develop a hybrid ranking algorithm combining vector similarity and hard-filter logic (e.g., "Must have 5 years experience") - Explanation Generation: Integrate a Local LLM to write a short summary justifying why a candidate fits or doesn't fit the role - API Design: Create FastAPI endpoints to upload files and retrieve ranked lists with scores - Visualization Data: Prepare aggregated data for potential frontend visualization (e.g., skill overlap charts) Requirements - Strong skills in Python, Pandas, and NumPy for data manipulation - Experience with NLP libraries. Knowledge of Vector Databases for similarity search - Experience with LLM prompting for Information Extraction - Ability to design RESTful APIs using FastAPI Nice to Have - Experience with OCR tools for handling scanned resumes - Knowledge of Docker for containerizing the parsing service - Familiarity with LangGraph to implement "Agentic Verification" Apply tot his job Apply To this Job