AI-powered mock interview platform that generates role-specific technical questions, evaluates answers in real-time, and tracks progress over time.
- Adaptive questions — generated based on your profile, target role, and past performance
- Code execution — write and run code in a sandboxed environment, validated against test cases
- Real-time evaluation — LLM scores answers on correctness, depth, and communication
- Progress tracking — per-topic scoring with round-by-round reports
- Text-to-speech — questions read aloud for realistic interview feel
- Research mode — scrape job postings to tailor questions to specific roles
- Backend: FastAPI + Pydantic
- LLM: Claude (Sonnet) for question generation and evaluation
- Code sandbox: Isolated subprocess execution with memory/time limits
- TTS: macOS
saycommand