AI Resume Screening App

This project was inspired by the endless scroll through resumes during campus placements. I wanted to build something that could quickly match candidates to job roles using AI. I used Python, Streamlit, and Flask to build the system, relying on NLP techniques like TF-IDF, spaCy NER, and cosine similarity to score resumes against job descriptions. I learned a lot about text preprocessing, model tuning, and dealing with messy real-world data. One of the biggest challenges was extracting clean data from PDFs and DOCX files—parsing resumes is messier than it sounds! But building something that automates such a painful task was totally worth it.

Built With

• cosine-similarity
• cursor
• github
• languages:-python-frameworks:-flask-(rest-api)
• ml-models)
• numpy
• pdfminer.six-(pdf-parsing)
• postman-(api-testing)
• python-docx-(docx-parsing)-storage:-csv-files-for-resume-and-job-data-tools:-git
• scikit-learn-(tf-idf
• streamlit-(ui)-nlp-&-ml:-spacy-(ner)
• trae
• transformers-(hugging-face)-data-handling:-pandas