VibeCheck.cv

Get Roasted. Get Hired.


VibeCheck.cv is a dual-mode professional identity agent designed to bridge the gap between "corporate fluff" and "high-impact branding." Built for the AI Hackathon, it uses a mix of brutal honesty and world-class optimization to help developers and creators own their digital identity on the Ola.cv ecosystem.

Inspiration

Most LinkedIn bios are 10% skills and 90% clichés. We wanted to build something that uses the viral nature of "AI Roasts" to grab attention, but provides immediate, tangible value by generating a deployable professional identity.

What it does

The application operates in three distinct phases:

  1. The Terminal Input: Users paste their current bio into a retro-styled cyberpunk interface.
  2. The Roast: Using Siray.ai/CometAPI and ElevenLabs, the AI analyzes the text and delivers a savage, audible critique of your buzzwords.
  3. The Glow-Up: The agent pivots, rewriting the bio into a high-performance narrative and extracting structured data for a .cv profile.

Tech Stack

  • Framework: Next.js 15 (App Router)
  • Infrastructure: Daytona (Standardized Cloud Development Environments)
  • Styling: Tailwind CSS v4 & Shadcn/UI
  • AI/LLM: OpenAI SDK (compatible with Siray.ai/CometAPI)
  • Voice: ElevenLabs API for real-time sarcastic audio
  • Identity: Ola.cv (Schema-ready JSON export)

Challenges we ran into

Implementing Tailwind CSS v4 alongside Shadcn/UI required manual adjustments to CSS variable mapping. Additionally, managing low-latency audio streaming for the ElevenLabs roast while the LLM was still generating text required a sophisticated streaming architecture.

Accomplishments that we're proud of

We successfully integrated a "dual-personality" system prompt that maintains state between a "Toxic VC" and a "Career Coach." Seeing the transition from a glitchy, red terminal to a clean, professional portfolio view is the highlight of the user experience.

What we learned

We learned the power of Vibe Coding—using AI agents to scaffold complex Tailwind layouts and handle boilerplate API routes allowed us to focus 100% on the "personality" of the agent.

The Math Behind the Vibe

To ensure the "Glow-Up" isn't just more fluff, we use a custom scoring algorithm to calculate the Impact Density \( ID \) of the new bio:

$$ID = \frac{\sum (Quantifiable Metrics)}{\text{Total Word Count}}$$

Where a higher \( ID \) suggests a more professional and less "mid" identity.

How to run it

# Clone the repo
git clone [https://github.com/shlokarth911/vibe_check.cv](https://github.com/shlokarth911/vibe_check.cv)

# Install dependencies
npm install

# Start the vibe
npm run dev

Built With

Share this project:

Updates