Description
The interview session completed successfully, but the automatic evaluation/report generation failed when calling the Gemini API.
The Rust LiveKit runner ended with interview_complete, but the report generation request returned 503 Service Unavailable.
As a result, no evaluation scores were produced and the report was marked as INCOMPLETE.
Error message
The automatic evaluation could not be completed: Rust LiveKit runner ended (interview_complete) but Gemini report generation failed.
Error: HTTP status server error (503 Service Unavailable) for url:
https://generativelanguage.googleapis.com/v1beta/models/gemini-3.1-flash-lite:generateContent
Your session ran end-to-end, but no scores were produced, so nothing here is an assessment of your work.
Check the agent logs and GOOGLE_API_KEY, then try again.
Expected behavior
After the interview reaches interview_complete, the system should generate the evaluation report and scores successfully.
Actual behavior
The interview completed normally, but report generation failed with HTTP 503 Service Unavailable, resulting in no scores being generated.
Possible improvement
It may be helpful to add automatic retry handling for temporary Gemini API 5xx errors and allow report generation to be retried without rerunning the full interview.
Description
The interview session completed successfully, but the automatic evaluation/report generation failed when calling the Gemini API.
The Rust LiveKit runner ended with interview_complete, but the report generation request returned 503 Service Unavailable.
As a result, no evaluation scores were produced and the report was marked as INCOMPLETE.
Error message
The automatic evaluation could not be completed: Rust LiveKit runner ended (interview_complete) but Gemini report generation failed.
Error: HTTP status server error (503 Service Unavailable) for url:
https://generativelanguage.googleapis.com/v1beta/models/gemini-3.1-flash-lite:generateContent
Your session ran end-to-end, but no scores were produced, so nothing here is an assessment of your work.
Check the agent logs and GOOGLE_API_KEY, then try again.
Expected behavior
After the interview reaches interview_complete, the system should generate the evaluation report and scores successfully.
Actual behavior
The interview completed normally, but report generation failed with HTTP 503 Service Unavailable, resulting in no scores being generated.
Possible improvement
It may be helpful to add automatic retry handling for temporary Gemini API 5xx errors and allow report generation to be retried without rerunning the full interview.