AI Interviewerv1.0
DashboardInterviews
Connected to the runtime worker via LiveKit.
Interviews
VT

Voice Test

Backend Engineer · Senior

test_voice_check_001

Completed

AI analysis

5/ 100
Strong No Hire

The candidate provided extremely short, fragmented, and largely incoherent responses to basic questions about a Python project they claimed to have built. There is no demonstrable evidence of Python proficiency, system design thinking, or senior-level engineering experience. The interview effectively did not progress past introductory questions.

Red flags
  • Extreme evasion of specifics across every probe: when asked about Python stack, async approach, and state management, candidate gave 2-3 word answers ('First API', 'entropic analysis') with no project detail at [45.7s], [88.6s], [113.4s].
  • Pattern of long pauses (8.0s, 8.7s, 8.2s, 10.8s, 11.6s) before extremely short, vague answers — inconsistent with someone recalling their own recent project. A real builder would speak fluently about their own system.
  • Factual confusion about own tooling: 'entropic analysis' / 'cloud entropy AI' at [45.7s] and [66.2s] when referring to Anthropic — implausible for someone who actually integrated Claude into a production voice agent.
Notable moments
Candidate·00:45

“We have used the entropic analysis.”

Garbled reference to Anthropic — suggests unfamiliarity with the tool they claim to have used in their own project.

Candidate·01:28

“Yes, basically I am managing this in passing type of communications.”

After a 10.8s pause, response is vague and grammatically broken, with no technical substance about conversation state or async handling.

Recording

Recording will be available after the interview ends.

Details

Duration
2m 24s / target 5m
Started
23 minutes ago
Ended
20 minutes ago
Outcome
Candidate disconnected

Transcript

19 turns
  1. C
    Candidate00:05

    Hello

  2. I
    Interviewer00:22

    Hey, good to meet you! Ready to dive in when you are. So let's start with Python since that's the core of this role. Tell me about a Python project you've worked on recently — something meaty, where you really had to think hard about how you structured things.

  3. C
    Candidate00:30

    Basically

  4. C
    Candidate00:30
Analyzed
20 minutes ago
Room
interview-test_voice_check_001
Required skills
Python
Request ID
045f59117ab143be9bdcb806d7cb0905

Created AI Voice Agent

  • I
    Interviewer00:39

    Oh nice, those are fun to build — lots of moving parts. What was the core of it, like what was the Python stack underneath?

  • C
    Candidate00:45

    We have used the entropic analysis.

  • I
    Interviewer00:58

    Interesting — sorry, did you say "entropic" analysis, or were you referring to Anthropic, like the Claude API?

  • C
    Candidate01:06

    I'm referring to cloud entropy AI.

  • I
    Interviewer01:17

    Got it, so Anthropic's Claude. Cool. What did the Python side look like — were you doing async stuff, managing conversation state, that kind of thing?

  • C
    Candidate01:28

    Yes, basically I am managing this in passing type of communications.

  • I
    Interviewer01:41

    Right, so like passing messages back and forth. How did you handle the async side of it — were you using asyncio, or something like FastAPI under the hood?

  • C
    Candidate01:53

    First API.

  • I
    Interviewer02:04

    Nice, FastAPI's a solid choice for that. So when you're managing conversation state across those voice turns — where were you actually

  • C
    Candidate02:04

    First Step

  • I
    Interviewer02:09

    Sorry, can you say that again? I think it cut off a bit.

  • C
    Candidate02:12

    Sorry

  • I
    Interviewer02:17

    No worries!

  • C
    Candidate02:19

    Okay, please repeat it.

  • C
    Candidate02:20

    Veera Supriyo