Cerebro Dashboard

A Sales Intelligence OS built with Next.js 14, Tailwind CSS, and Perplexity AI.

Features

• Analysis Page: Upload files or paste text to extract customer pain points using AI
• Product Vault: Manage product features and capabilities (coming soon)
• Radar: Monitor customer needs and opportunities (coming soon)

Getting Started

1. Install dependencies:

npm install

2. Set up environment variables:

cp .env.example .env

3. Add your Perplexity API key to .env:

PERPLEXITY_API_KEY=your_perplexity_api_key_here

Get your API key from Perplexity Settings

4. Run the development server:

npm run dev

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

How It Works

The Insight Engine (Pillar 1: Listen)

The /api/analyze route uses Perplexity AI to extract structured customer needs from transcripts:

• Pain Point: The specific problem or need mentioned
• Urgency: A score from 1-5 (1 = very low, 5 = critical)
• Mentioned Timeline: Any timeline mentioned (e.g., "Q3", "next quarter", "in 6 months")

Simply paste a transcript or upload a text file, and the AI will automatically extract and display customer pain points as cards.

Tech Stack

• Frontend: Next.js 14 (App Router) + Tailwind CSS
• AI Engine: Perplexity API (llama-3.1-sonar-small-128k-online)
• Database: Supabase (configured, ready for future use)
• Icons: Lucide React
• Hosting: Vercel-ready

Project Structure

/app
  /api
    /analyze          # Perplexity AI-powered analysis endpoint (returns JSON only)
    /analyze-and-save # Analyzes transcript and saves to Supabase database
  /page.tsx           # Analysis page (main)
  /product-vault      # Product Vault page
  /radar              # Radar page
/components
  Sidebar.tsx         # Navigation sidebar
  PainPointCard.tsx   # Pain point display component
/lib
  analyze-transcript.ts # Core function to analyze and save insights
  supabase.ts          # Supabase client configuration

Database Integration

Analyze and Save Function

The analyzeTranscriptAndSave function processes transcripts and automatically saves insights to your Supabase database:

import { analyzeTranscriptAndSave } from "@/lib/analyze-transcript";

const insights = await analyzeTranscriptAndSave(transcript, callId);

Or use the API route:

POST /api/analyze-and-save
{
  "transcript": "Customer call transcript...",
  "call_id": "uuid-from-calls-table"
}

See /lib/README.md for detailed usage and database schema requirements.