This project is an end-to-end Generative AI application that analyzes a candidate’s CV against a target job role, identifies skill gaps, and generates personalized learning insights using:
- Rule-based skill extraction with evidence
- Role-scoped knowledge (core / optional / excluded skills)
- Retrieval-Augmented Generation (RAG)
- Vector databases (ChromaDB)
- Large Language Models (Groq)
- Streamlit UI
The system is role-aware, grounded, and production-ready, with graceful fallback when LLMs are unavailable.
- Upload CV in PDF or DOCX
- Regex-based skill extraction with evidence
- Role-scoped filtering of required skills
- RAG pipeline using ChromaDB
- Learning guidance via Playbooks
- Career grounding via Roadmap documents
- LLM-generated professional insights (optional)
- Fault-tolerant LLM layer (UI never crashes)
- Deployable on Streamlit Cloud
pip install -r requirements.txt
streamlit run app.py