Roleplay RAG

Overview

Roleplay RAG lets you rehearse high-stakes conversations. The AI always plays the authority figure, such as the interviewer, visa officer or doctor, and you play the other role. Upload a document like your resume and it asks questions about your actual content instead of generic ones. Built as a college project.

Key Features

• Nine Scenarios: Job interview, visa interview, doctor and patient, viva voce, college admission, traffic stop, landlord and tenant, performance review, and bank loan interview.
• Grounded in Your Documents: Upload a PDF or TXT, or paste details, and the AI retrieves relevant passages each turn to ask specific follow-ups.
• Setting and Tone: Choose a location and a mood from Relaxed to Strict, or define your own.
• Session Feedback: End the session for an evaluation of your performance with strengths, areas to improve, a rough rating, and tips.
• Local or Cloud Models: Use Ollama, LM Studio, llama.cpp or vLLM locally, or OpenAI, Anthropic, Gemini, Grok, Groq and MiniMax with your own key.
• Private by Design: Everything stays in memory for the session, and API keys are never written to disk.

"Practice the hard conversation before it counts."