How pranavkrishnasuresh/resumeprocessing works
Main Components
• Input Source – a Resume PDF input module that will eventually connect to the Ceipal database.
• Pre-processing Steps – a module (or multiple sub-components) handling PDF-to-text conversion.
• Resume Summarization and Categorization – this includes the initial summary generation and an iterative categorization step.
• Scoring Module – where the resume is scored using the Baseline Meta LLama3 model (to later be replaced with a fine-tuned model).
• Interview Question Generation – a component that generates five personalized interview questions if the resume score meets a threshold.
• External Integrations – referencing future integrations with the Ceipal database (for resume fetching) and the post function to transfer scored resume data to a database.
File Structure to Reflect Modules
• The file “resumeScoreUpdated.py” likely contains the resume scoring logic and perhaps the overall orchestration of the pipeline.
• The file “skills_compare.py” might be responsible for comparing the resume’s skills to some baseline or performing a part of the categorization.
• The requirements.txt indicates the Python packages used by these scripts, and although not a module in itself, it gives insight into the external dependencies.
• From the scoring module, if the score is above a threshold, branch out to the “Interview Question Generator” node.