Resume Insights is an advanced resume parsing and analysis tool that leverages the power of LlamaIndex, Gemini, and Streamlit to extract key information from PDF resumes and present it in an intuitive, interactive format.
- PDF Resume Parsing: Seamlessly extract information from PDF resumes using LlamaParse.
- AI-Powered Analysis: Utilizes Gemini LLM for intelligent information extraction and skill matching.
- Interactive UI: Built with Streamlit for a user-friendly experience.
- Skill Visualization: Displays top skills with interactive progress bars.
- Job Matching: Analyzes skill relevance for specific job positions.
- Structured Output: Extracts name, email, age, and skills in a structured format using Pydantic models.
-
Clone the repository:
git clone https://github.com/luillyfe/resume-insights.git cd resume-insights -
Install dependencies:
pip install -r requirements.txt -
Set up environment variables:
export GOOGLE_API_KEY=your_google_api_key export LLAMA_CLOUD_API_KEY=your_llama_cloud_api_key -
Run the Streamlit app:
streamlit run app.py -
Open your browser and navigate to
http://localhost:8501.
- LlamaIndex: For document processing, indexing, and querying.
- Gemini: AI model for natural language understanding and generation.
- Streamlit: For building the interactive web application.
- Pydantic: For data validation and settings management.
- LlamaParse: For efficient PDF parsing.
- Document Processing: Resumes are parsed using LlamaParse and split into manageable chunks.
- Indexing: Processed documents are indexed using LlamaIndex's VectorStoreIndex.
- Information Extraction: The system extracts key information like name, email, age, and skills.
- Skill Analysis: Top skills are displayed with visualizations.
- Job Matching: Skills are analyzed for relevance to specific job positions.
- Interactive UI: Users can explore extracted information and job matching results through a Streamlit interface.
ResumeInsights: Main class for processing resumes and extracting information.Candidate: Pydantic model for structured candidate data.JobSkill: Pydantic model for skill relevance to job positions.- Streamlit app: Provides the user interface for uploading resumes and displaying insights.
- The free tier of the Gemini API/Google Cloud has a 10k byte limit on request payloads, which may limit the number of skills that can be processed: issue.
- LlamaIndex may face challenges with knowledge dispersed across different sections of a document, such as inferring age or skill proficiency from work experience.
Contributions are welcome! Please feel free to submit a Pull Request.
This project is licensed under the MIT License - see the LICENSE file for details.
For any queries or suggestions, please open an issue or contact luillyfe89@gmail.com.
Made with ❤️ by Fermin Blanco