SpeechStack
Submit a template
← All templatestemplates / technical-interview-practice-agent-with-real-time-feedback
Cartesiaupdated May 23, 2025 · education · interview-practice

Technical Interview Practice Agent with Real-Time Feedback

An AI interviewer that conducts technical job interviews with real-time parallel analysis by three specialized evaluation agents. The system provides structured feedback on technical expertise, communication skills, and logical reasoning, helping candidates practice and improve their interview performance through voice-based interactions.

Try the demoView sourceFork template
The numbers
latency
cost / min
frameworkCartesia
The stack
telephonyWeb Only
speech-to-textCartesia STT
llmCerebras Llama 3.3 70B
text-to-speechCartesia Sonic-3
System prompt
No prompt published.
Config
config.json
{
  "tools": [
    {
      "name": "start_interview",
      "description": "Initiates the interview session after user confirms readiness"
    },
    {
      "name": "end_call",
      "description": "Terminates the interview session"
    }
  ],
  "architecture": "multi-agent",
  "response_format": "structured_output",
  "main_agent_model": "cerebras/llama3.3-70b",
  "background_judges": [
    {
      "name": "technical_report",
      "description": "Evaluates technical expertise"
    },
    {
      "name": "communication_report",
      "description": "Evaluates communication skills"
    },
    {
      "name": "reasoning_report",
      "description": "Evaluates logical thinking"
    }
  ],
  "judge_agent_model": "cerebras/llama3.1-8b"
}
Tools
  • start_interviewInitiates the interview session after user confirms readiness
  • end_callTerminates the interview session
Tags
multi-agentinterview-prepstructured-feedbackbackground-analysistechnical-interview
Voice Notes

Voice AI recipes, picks, and analysis.

Get the useful new templates plus the occasional teardown of what’s working in production voice AI.

contributed by @lucyliulee · MIT · source: github discoverylanguages: en-US