Use a voice agent for oral exams

A professor used a voice agent to run oral finals, then had an LLM panel grade the transcripts and flag shared weak spots.

Done with Custom agent· Advanced· Jan 2026

What they did
They used a voice agent to interview students about project work and capture a transcript of each exam. The setup also included a group of LLMs that reviewed the transcripts, talked through the grading, and produced an audit trail. In all, the process covered 36 students across 9 days, with each interview lasting about 25 minutes.
How it went
The whole process cost $15, or about $0.42 per student. It also revealed a class-wide gap in A/B testing, and 13% of students said they preferred the AI-led oral format.
Tips
Build in transcript review and an audit trail so you can check the grading later. Keep a human approval step before any scores or feedback are sent to students.

Try it yourself with Gemini

I’m running oral exams for [course name] with [number] students on [project/topic]. Create a voice-interview script with [number] questions that covers [topics], asks clear follow-ups when answers are vague, and saves a transcript for each student. Then review each transcript against this rubric: [rubric]. Before any grades, feedback, or results are sent to students, pause and show me the draft scoring and the class-wide patterns you found for approval.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on x.com ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.