Built using BERT-based semantic embeddings, cosine similarity ranking, OCR-based resume parsing, and skill gap analysis — transforming resumes into actionable hiring insights.
Resumes categorized across AI, Data Science, .NET, and multiple tech domains. Includes scanned PDFs processed via OCR.
JSON-based job dataset including skills, responsibilities, experience level, and domain-specific keywords.
OCR → Text Cleaning → SentenceTransformer Embedding → Cosine Similarity → Top-K Ranking → Skill Gap Analysis.
Implemented lexical similarity baseline using TF-IDF vectorization and cosine similarity for benchmarking.
Dense semantic embeddings improved contextual matching accuracy, capturing meaning beyond keyword overlap.
Upload your resume and receive top job matches with confidence score and skill insights.