Can I vibecode Jamie?

text
Build a deliberately narrow personal substitute for Jamie, focusing on privacy-first, bot-free meeting capture and intelligence.
Use exactly this stack: Python 3.12 + FastAPI + whisper.cpp + SQLite for the backend, and a lightweight Electron or Tauri wrapper for the native desktop feel.
Primary job: Build a private, native-feeling desktop app that captures system audio directly (no meeting bots joining calls), transcribes it locally, and generates executive summaries.
Start from an empty folder and create the complete working project. Make the default mode single-user, private, and fully offline-capable for transcription.
Store all user data locally on the machine. Do not add analytics, telemetry, ads, or third-party accounts. Emphasize a strict "privacy-first" design where data never leaves the device unless an external LLM API is explicitly configured.
Put every secret and external credential in .env and provide .env.example.
Implement an "Executive Assistant Sidebar" UI: users should see the meeting notes generating, and be able to use a chat interface to ask questions about the transcript (e.g., "What was the final deadline?").
Generate structured notes that include high-level summaries, key decisions, and action items extracted from the transcript.
Implement the smallest polished interface that completes the core loop end to end, supporting both online meetings (capturing system audio) and in-person discussions (capturing microphone audio).
Include clear empty, loading, validation, success, and failure states. Add Markdown export so the user is not trapped in the app.
Use accessible keyboard navigation, labels, focus states, and sensible contrast.
Validate untrusted input and never log secrets or private file contents.
Deliberately exclude these paid-product advantages: calendar and CRM integrations, cross-call team analytics, cloud syncing, and multi-device coordination.
Do not fake integrations, network effects, proprietary data, model quality, compliance, or security claims.
Where an external API (like OpenAI for summarization) is optional, keep the app useful without it and clearly explain the degraded local-only mode.
Write focused unit tests for the audio capture logic, data model, and transcription workflow. Add one end-to-end smoke test.
Create a README with setup, permissions (especially OS-level audio capture permissions), architecture, data location, backup, and limitations.
Add scripts for install, development, test, build, and a production-style local run.
Run the tests and build before finishing, then fix errors rather than merely describing them.