An intelligent, dual-purpose application designed to optimize the job search process. OpenGuild combines a RAG-powered Resume Analyzer to evaluate candidate fit, an integrated Job Application Tracker to manage the hiring pipeline, and a Voice-Driven AI Mock Interviewer to help you prep for the big day.
Built with Python, Streamlit, LangChain, FAISS, and OpenAI, and fully containerized with Docker.
(Note: Ensure your architecture images are uploaded to your repo to display correctly)
- Semantic Match Scoring: Uses vector embeddings (FAISS) to calculate a 0-100% quantitative fit score based on hard skills, experience, and industry context.
- Smart PDF Parsing: Implements layout-aware document chunking using PyMuPDF and pdfplumber to accurately read complex, multi-column resumes without losing context.
- Comprehensive SWOT Analysis: Automatically generates Strengths, Weaknesses, Opportunities, and Threats for the candidate relative to the specific role.
- Elevate Speech & Application Kit: Drafts a tailored cover letter and helps you structure answers using industry frameworks like STAR (Situation, Task, Action, Result) and HERO (Headline, Effect, Rationale, Operation).
- Exportable Reports: Download the full analysis as a formatted Markdown file.
- Seamless Integration: One-click save from the Analyzer directly to your Tracker, auto-extracting the Company Name and Job Title using structured LLM outputs.
- Visual Dashboard: Real-time metrics and charts displaying pipeline health, interview statuses, and application momentum over time.
- Interactive Data Editor: Update application statuses (e.g., "Applied" -> "Interviewing") directly within the UI.
- Persistent Storage: Data is saved locally via CSV, ensuring your pipeline survives container restarts.
- Practice Under Pressure: Do a live audio interview with your AI Coach and get real-time evaluations on your delivery.
- Tailored to the Role: The AI automatically adopts the persona of the hiring manager for the specific job and company you saved in your tracker.
- Speech-to-Text & Neural Voice: Speak naturally into your microphone. The AI transcribes your answer, processes it, and speaks back to you in a realistic neural voice.
- Performance Report: Get personalized, expert feedback highlighting where you did well and where you can improve before the real interview.
- Frontend: Streamlit
- AI & Orchestration: LangChain Core, OpenAI API (
gpt-4o) - Audio Processing: OpenAI Whisper (STT) and TTS-1 (Text-to-Speech)
- Vector Database: FAISS (Local)
- Data Processing: Pandas, PyMuPDF, pdfplumber
- Deployment: Docker & Docker Compose
- Python 3.13+ (for local development)
- Docker and Docker Compose (for containerized deployment)
- An active OpenAI API Key
⚠️ Security Note: Never commit your.envfile to GitHub. It is already included in the.gitignore.
git clone [https://github.com/NimdaGrogu/job-hunt-assistant.git](https://github.com/NimdaGrogu/job-hunt-assistant.git)
cd job-hunt-assistant
touch src/job_tracker.csvCreate a file named .env in the src/ folder (or project root) and add your keys:
OPENAI_API_KEY=sk-your-actual-api-key-here
VERBOSE_RAG_LOGS=false (Note: When VERBOSE_RAG_LOGS=true, detailed LangChain callbacks will print in the terminal for debugging).
pip install --upgrade pip
pip install -r requirements.txtstreamlit run src/app.pyThe app should open automatically at http://localhost:8501
Build the Image:
docker build -t openguild .Run the Container (Passes your local .env file securely):
docker run -p 8501:8501 --env-file src/.env openguildAssuming you are in the root directory and your .env is configured:
docker compose up -dAccess the app by opening your browser to http://localhost:8501
⚖️ LLM as Judge (WIP): Integrating the Gemini API to act as a secondary guardrail, ensuring fair and unbiased evaluation of candidates regardless of background, ethnicity, or socio-economic position.
Pull requests are welcome! For major changes, please open an issue first to discuss what you would like to change.
MIT