Step 2: Connect Bright Data API and Snapshot Polling

Trigger a Bright Data dataset run, then poll for completion and retrieve the snapshot when ready.

1. Configure Bright Data Trigger Request with URL https://api.brightdata.com/datasets/v3/trigger and Method POST.
2. Set JSON Body to =[ { "location": "{{ $json['Job Location'] }}", "keyword": "{{ $json.Keyword }}", "country": "{{ $json['Country (2 letters)'] }}" } ] .
3. In Query Parameters, set dataset_id to [YOUR_ID], include_errors to true, type to discover_new, discover_by to keyword, and uncompressed_webhook to true.
4. Add a header in Bright Data Trigger Request: Authorization set to Bearer [CONFIGURE_YOUR_TOKEN].
5. Set Delay For Snapshot Poll to Unit minutes to wait between status checks.
6. Set Check Snapshot Status URL to =https://api.brightdata.com/datasets/v3/progress/{{ $('Bright Data Trigger Request').item.json.snapshot_id }} and include the same Authorization header.
7. Configure Snapshot Ready Decision with condition Left Value ={{ $json.status }} and Right Value running to loop back via Delay For Snapshot Poll until the snapshot is ready.
8. Set Retrieve Snapshot Data URL to =https://api.brightdata.com/datasets/v3/snapshot/{{ $('Bright Data Trigger Request').item.json.snapshot_id }} and set query parameter format to json.

Step 3: Connect Google Sheets

Store retrieved job data and later update each row with the generated pitch.

1. In Append Job Listings Sheet, set Operation to append, Document ID to [YOUR_ID], and Sheet Name to gid=0.
2. Credential Required: Connect your googleSheetsOAuth2Api credentials in Append Job Listings Sheet.
3. In Update Pitch Column, set Operation to update, Document ID to [YOUR_ID], and Sheet Name to gid=0.
4. Map Pitch to ={{ $json.text }} and company_name to ={{ $('Split Job Records').item.json.company_name }}, with Matching Columns set to company_name.
5. Credential Required: Connect your googleSheetsOAuth2Api credentials in Update Pitch Column.

Step 4: Prepare Job Records for AI Processing

Split the appended dataset into individual records before generating pitches.

1. Configure Split Job Records with Field to Split Out set to company_name, job_title, description_text.
2. Verify the connection flow Append Job Listings Sheet → Split Job Records → LLM Pitch Generator to ensure each job record is handled independently.

Step 5: Set Up LLM Pitch Generator

Generate a concise pitch or mark irrelevant roles using the connected OpenAI model.

1. In LLM Pitch Generator, set Prompt Type to define and Text to the full expression-driven prompt starting with =Read these fields from the job post: - Company: `{{ $json.company_name }}` - Title: `{{ $json.job_title }}` - Description: `{{ $('Append Job Listings Sheet').item.json.job_overview }}`.
2. Confirm OpenAI Chat Engine is connected to LLM Pitch Generator as the language model and set the model to gpt-4o-mini.
3. Credential Required: Connect your openAiApi credentials in OpenAI Chat Engine (credentials are applied to the parent model, not inside LLM Pitch Generator).

Step 6: Test and Activate Your Workflow

Run the workflow end-to-end to confirm job data is collected, processed, and updated in Sheets.

1. Click Execute Workflow and submit the Job Search Form Trigger form with sample data.
2. Watch the polling loop: Delay For Snapshot Poll → Check Snapshot Status → Snapshot Ready Decision until the snapshot is retrieved.
3. Confirm rows appear in the sheet after Append Job Listings Sheet and that Update Pitch Column fills the Pitch field.
4. When the results look correct, toggle the workflow to Active for production use.