OneOrigin interview workspace for candidate sessions, AI-assisted collaboration, and recruiter reports.

Getting Started

Install dependencies:

npm install

Create a .env file with:

DATABASE_URL="file:./dev.db"
OPENAI_API_KEY="your-key"
OPENAI_MODEL="gpt-5"

Run the development server:

npm run dev

Open http://localhost:3000 in your browser.

Product Overview

• Candidate dashboard with multi-round interview flow.
• Mode-specific prompts:
  • Build & ship: production-ready feature, tests, deployment notes.
  • Debug & refactor: regression diagnosis, fix, reliability safeguards.
  • Design & align: architecture, tradeoffs, stakeholder alignment.
• AI assistant that only helps when prompted and avoids full solutions.
• Proctoring: tab/window switches are recorded, warned to candidates, and included in reports.
• Reports dashboard with scores, highlights, and notes.

Routes

• "/": interview creator (role/level/mode + share link).
• "/candidate?t=TOKEN": candidate workspace.
• "/reports": recruiter scorecards.

Environment Variables

• DATABASE_URL: Prisma SQLite DB (e.g. file:./dev.db).
• OPENAI_API_KEY: enables AI assistant + report evaluator.
• OPENAI_MODEL: model name (defaults to gpt-5).

Notes

• If OPENAI_API_KEY is missing, report evaluation falls back to a local heuristic.
• Proctoring events are client-side signals; treat as advisory.