3. Event Attendee scraper

## 🤖 **SECTION 2: Smart Scraping via AI Agent**

This is where the **heavy lifting** happens using Bright Data’s **Mobile Carrier Proxy (MCP)** and OpenAI’s logic to bypass website protections.

---

### 🧠 `🤖 Bright Data AI Agent`

This node coordinates scraping. It tells the agent exactly:

* Which URL to open
* What to extract (like attendees, venues, events)
* How to format the response (in markdown or JSON)

> ⚡ *Why it's powerful:* The AI Agent understands human-like prompts and adapts scraping automatically.

---

#### 🧠 `🧠 Agent Memory Model`

Helps the Agent remember its instructions or carry session context (e.g., "continue scrolling", "grab the next batch").

---

#### 🌐 `🕷️ MCP Scraper: 10times`

This is the **actual web scraper** running on **Bright Data’s mobile proxy** — which mimics a real mobile device to avoid bot detection. It loads JavaScript, scrolls pages, and handles dynamic content.

> 🛡️ *Biggest Advantage:* Can access pages that block most traditional scrapers!

---

#### 📄 `🧾 Parse Attendee Data`

Uses the **Structured Output Parser** to turn messy scraped markdown into clean, structured JSON with fields like:

* `categories`
* `featured_events`
* `attendee_feedback`
* `venues`

> 🧩 *Tip:* This is the secret to clean automation — parsed data can now be reused, filtered, or enriched!

---