This tutorial will show you how to build a simple “signup radar” that takes an email address and returns information about the person who signed up, using the Linkup API’s structured output feature.

​
What We’re Building

Our signup radar will:

  • Take an email address as input
  • Use Linkup API to search for information about the person
  • Return structured data about the person (name, position, company, LinkedIn URL, etc.)

​
Prerequisites

  • A Linkup API key
  • Python or Node.js installed

Get your API key

Create a Linkup account for free to get your API key.

​
Step 1: Install the SDK

pip install linkup-sdk

​
Step 2: Set Up the Client

from linkup import LinkupClient

client = LinkupClient(api_key="<YOUR_LINKUP_API_KEY>")

​
Step 3: Define the Structured Output Schema

The key to our signup radar is using Linkup’s structured output feature. We need to define a schema that specifies what information we want to extract.

import json

schema = {
  "type": "object",
  "properties": {
    "fullName": {
      "type": "string",
      "description": "The full name of the person"
    },
    "company": {
      "type": "string",
      "description": "The company the person works for"
    },
    "position": {
      "type": "string",
      "description": "The job title or position of the person"
    },
    "linkedInUrl": {
      "type": "string",
      "description": "The LinkedIn profile URL of the person"
    },
    "companyWebsite": {
      "type": "string",
      "description": "The website of the company"
    },
    "additionalInfo": {
      "type": "string",
      "description": "Any additional relevant information about the person"
    }
  },
  "required": ["fullName", "company"]
}

schema_str = json.dumps(schema)

​
Step 4: Create the Signup Radar Function

def signup_radar(email):
    # Extract name and domain
    name_part = email.split('@')[0]
    domain = email.split('@')[1]
    
    # Format name for searching (convert saksena to Saksena)
    formatted_name = name_part.capitalize()
    
    # Determine company from domain (if not common email provider)
    company_hint = ""
    common_domains = ["gmail.com", "hotmail.com", "outlook.com", "yahoo.com", "icloud.com"]
    if domain not in common_domains:
        company_hint = domain.split('.')[0]
    
    # Create search query
    if company_hint:
        query = f"Find the LinkedIn profile URL for {formatted_name} who works at {company_hint}. Return their full name, current position, and company information."
    else:
        query = f"Find the LinkedIn profile URL for someone with the email username {formatted_name}. Return their full name, current position, and company information."
    
    # Call Linkup API
    response = client.search(
        query=query,
        depth="deep",  # Use deep for more thorough results
        output_type="structured",
        structured_output_schema=schema_str
    )
    
    return response

# Example usage
from pprint import pprint

result = signup_radar("philippe@linkup.so")
pprint(result)

​
Step 5: Enhanced Query Generation

Let’s improve our query to get better results:

def generate_query(email):
    name_part = email.split('@')[0]
    domain = email.split('@')[1]
    
    # Handle different name formats (snake_case, dot.case, etc.)
    if "_" in name_part:
        name_parts = name_part.split("_")
        formatted_name = " ".join(part.capitalize() for part in name_parts)
    elif "." in name_part:
        name_parts = name_part.split(".")
        formatted_name = " ".join(part.capitalize() for part in name_parts)
    else:
        formatted_name = name_part.capitalize()
    
    # Determine company from domain
    common_domains = ["gmail.com", "hotmail.com", "outlook.com", "yahoo.com", "icloud.com"]
    if domain not in common_domains:
        company = domain.split('.')[0].capitalize()
        return f"Find the LinkedIn profile URL for this person: {formatted_name} at {company}. If the domain of the email address ({domain}) is not a common email provider, it is probably the name of the company this person works for and {domain} is probably the company website, so search specifically for someone with that name at this company. Return their full name, position, company details, LinkedIn URL, as well as any relevant information you can find about them."
    else:
        return f"Find the LinkedIn profile URL for this person with email username {formatted_name}. Return their full name, current position, company, and LinkedIn URL."

​
Full Implementation

Let’s put it all together:

from linkup import LinkupClient
import json
from pprint import pprint

class SignupRadar:
    def __init__(self, api_key):
        self.client = LinkupClient(api_key=api_key)
        self.schema = {
            "type": "object",
            "properties": {
                "fullName": {
                    "type": "string",
                    "description": "The full name of the person"
                },
                "company": {
                    "type": "string",
                    "description": "The company the person works for"
                },
                "position": {
                    "type": "string",
                    "description": "The job title or position of the person"
                },
                "linkedInUrl": {
                    "type": "string",
                    "description": "The LinkedIn profile URL of the person"
                },
                "companyWebsite": {
                    "type": "string",
                    "description": "The website of the company"
                },
                "additionalInfo": {
                    "type": "string",
                    "description": "Any additional relevant information about the person"
                }
            },
            "required": ["fullName", "company"]
        }
        self.schema_str = json.dumps(self.schema)
    
    def generate_query(self, email):
        name_part = email.split('@')[0]
        domain = email.split('@')[1]
        
        # Handle different name formats (snake_case, dot.case, etc.)
        if "_" in name_part:
            name_parts = name_part.split("_")
            formatted_name = " ".join(part.capitalize() for part in name_parts)
        elif "." in name_part:
            name_parts = name_part.split(".")
            formatted_name = " ".join(part.capitalize() for part in name_parts)
        else:
            formatted_name = name_part.capitalize()
        
        # Determine company from domain
        common_domains = ["gmail.com", "hotmail.com", "outlook.com", "yahoo.com", "icloud.com"]
        if domain not in common_domains:
            company = domain.split('.')[0].capitalize()
            return f"Find the LinkedIn profile URL for this person: {formatted_name} at {company}. If the domain of the email address ({domain}) is not a common email provider, it is probably the name of the company this person works for and {domain} is probably the company website, so search specifically for someone with that name at this company. Return their full name, position, company details, LinkedIn URL, as well as any relevant information you can find about them."
        else:
            return f"Find the LinkedIn profile URL for this person with email username {formatted_name}. Return their full name, current position, company, and LinkedIn URL."
    
    def lookup(self, email):
        query = self.generate_query(email)
        
        response = self.client.search(
            query=query,
            depth="deep",  # Use deep for more thorough results
            output_type="structured",
            structured_output_schema=self.schema_str
        )
        
        return response

# Example usage
if __name__ == "__main__":
    radar = SignupRadar(api_key="<YOUR_LINKUP_API_KEY>")
    
    # Example emails
    emails = [
        "philippe@linkup.so",
        "boris@linkup.so"
    ]
    
    for email in emails:
        print(f"\nLooking up: {email}")
        result = radar.lookup(email)
        pprint(result)

​
How It Works

  1. Email Analysis: The tool parses the email to extract the username and domain.
  2. Query Generation: It creates a smart search query based on the email components:
    • Formats the username to handle common patterns (first.last, first_last)
    • Uses the domain as a company hint if it’s not a common email provider
  3. Structured Output: Uses Linkup’s structured output feature with a custom schema to ensure consistent, well-formatted results.
  4. Deep Search: Uses the “deep” search depth for more comprehensive results.

​
Test Examples

Try the signup radar with these email examples:

​
Advanced Enhancements

For a production version, consider adding:

  • Other relevant information on your users you receive in the sign up form. These should be added to the prompt
  • Better manage ambiguity when multiple people could own the same email. Change the prompt and the structured output format to allow for multiple potential people
  • Error handling for invalid emails or API failures
  • Rate limiting to manage API usage
  • Async batch processing for multiple emails

​
Conclusion

You’ve now built a simple but powerful “signup radar” using the Linkup API that can extract structured information about users from just their email address. The structured output feature ensures you get consistent, well-formatted data that can be easily integrated into your systems.

Facing issues? Reach out to our engineering team at support@linkup.so

Structured Output GuidePrompt Catalog