This repository introduces a multi stage pipeline for skill extraction. The pipeline includes a curriculum bi-encoder, trained to retrieve relevant skill candidates for job description sentences. Then in Stage 2 a cross-encoder model is trained with a binary ranking objective, to determine which of the retrieved candidates are true.

The Personal Career Navigator project uses a large language model (LLM) – specifically OpenAI's GPT-3.5-turbo – as its core reasoning engine. This model powers every step of the agent’s multi‑stage pipeline: skill extraction, job requirement generation, gap analysis, and roadmap creation.