| Demo Script: HiredGPT | |
| DuelHiredGPT Duel – Competitive Interview Simulation | |
| Hello everyone, | |
| I’m Amal, and this is HiredGPT Duel – a competitive interview simulator that shows you exactly how you stack up against a top-tier candidate. | |
| 1 — Problem & Idea\ | |
| Most interview-prep tools grade you in isolation. You get a score, but no sense of what great really looks like. | |
| HiredGPT Duel fixes that by pitting you against an AI “star applicant” who’s about 10 percent stronger than your own résumé. After every answer you see a side-by-side comparison and immediate feedback. | |
| 2 — Quick Walk-through\ | |
| 1. Upload your resume and the target job description in pdf format or paste job description text format. | |
| 2. Pick how much stronger the rival should be—10 to 100 percent—or leave it at the default 10 percent. | |
| 3. Enter the job role you are targeting | |
| 4. Submit to generate rival resume and interview questions. | |
| 5. Behind the scenes, the system generates (LLM) interview questions based on submited resume JD and job description with few shot prompt engineering tech. | |
| 6. System generates difficulty level, job desciption, resume and key factors exteacted. | |
| Now we jump into the duel. | |
| - Question view | |
| on the top | |
| - Answer the question—type or speak. If you record audio, Whisper transcribes it instantly. | |
| - Click “Generate Rival Answer.” The AI competitor responds using the upgraded résumé. | |
| - | |
| 3 — Tech Highlights | |
| Frontend: Streamlit for a fast, sharable UI. | |
| LLMs: Google Gemini (via Groq client) for question generation, résumé enhancement, rival answers, and scoring and report generation. | |
| Speech-to-Text: OpenAI Whisper running locally for offline transcription | |
| Text-to-Speech: pyttsx3 python library. if you’d like the rival to read answers aloud. | |
| Document Handling: PyPDF2 for parsing PDFs. | |
| Everything runs on-device except the LLM calls, keeping latency low and privacy high. | |