Replace retired Vapi workflow with an inline generator assistant

[deya19]

Summary

Vapi retired Workflows on 2026-08-18, so vapi.start(NEXT_PUBLIC_VAPI_WORKFLOW_ID) now fails with a 400 ("Workflows were retired... Build with Assistants or Squads instead"). The "generate" flow is moved onto an inline assistant, and interview creation — previously done by the workflow's server step — is now driven from the client after the call.

• constants: new generator: CreateAssistantDTO that asks for role, level, techstack, focus, and question count, and ends the call itself via tools: [{ type: "endCall" }].
• Agent.tsx: vapi.start(generator, { variableValues: { username, userid } }); on call end in generate mode it POSTs the transcript to /api/vapi/generate before routing home.
• /api/vapi/generate: accepts the new transcript shape while staying backward compatible with explicit fields:

const { type, role, level, techstack, amount } = body.transcript
  ? await extractDetailsFromTranscript(body.transcript) // generateObject + zod schema, gemini-2.0-flash-001
  : body;

NEXT_PUBLIC_VAPI_WORKFLOW_ID is no longer used and can be dropped from the env.

Link to Devin session: https://app.devin.ai/sessions/2445170116a6436db60a4a4242adf506
Open in Devin Desktop: https://app.devin.ai/desktop/session/2445170116a6436db60a4a4242adf506?variant=devin

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
interview-mock-platform	Error		Sep 12, 2026 1:50pm UTC

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: aa81a8a0-46f7-463c-af40-07c0132e0c8a

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}