GrillKit – self-hosted AI technical interview trainer with voice (github.com)

GrillKit has launched an open-source, self-hosted AI technical interview trainer that allows users to configure OpenAI-compatible models and practice with real-time scoring. Users can set up interviews using YAML question banks tailored to various programming languages and experience levels, while also utilizing features like optional voice input and audio for questions. The user-friendly dashboard offers quick access to recent sessions and streamlined interview setup options, enhancing the training experience with real-time Q&A scoring and feedback. This development is significant for the AI/ML community as it democratizes access to sophisticated interview training tools, enabling a wider audience to refine their skills in a highly interactive format. Key technical features include multi-language support, an integrated WebSocket for real-time interactions, and the ability to incorporate voice responses via Whisper dictation and Piper text-to-speech for audio questions. Future enhancements will include additional programming question banks and a code editor, making GrillKit an increasingly versatile platform for aspiring tech professionals.