An automated platform conducting realistic developer technical interviews with instant AI evaluation and resume feedback.
This capstone project simulates real-world developer hiring rounds. It generates contextual technical questions based on the candidate's resume, records audio/text responses, and provides comprehensive feedback on code performance, communication skills, and architectural knowledge using the Gemini API.
Integrating the streaming voice response with the Gemini completion API while keeping response latency under 1 second.
Designed a buffer-caching middleware that batches transcriptions locally before dispatching to the LLM agent model.
Implemented real-time speech-to-text response capturing.
Created custom evaluation algorithm tracking correctness, formatting, and optimization suggestions.
Designed responsive dashboard showing progress analytics over multiple mock trials.
Learn the exact tools, architectures, and programming systems used by Rohan Sharma to deploy this project. Join our next live career cohort.