I recently built a Make.com scenario that watches Gmail for job alerts from LinkedIn and Freelancer.com, extracts the job description, and uses GPT-4o-mini to automatically write a tailored proposal — all saved to Google Sheets.
A few things I learned along the way:
- HTML to Text parser is essential before sending to OpenAI
- Setting Max Tokens to 800 prevents cut-off responses
- Using “Contains” filter for sender email is more reliable than exact match
Happy to share more details if anyone is interested!
Following up on my post yesterday, a few people sent DMs asking what the final output looks like.
Here is a quick screenshot of the exact Make scenario running in the background and how it structures the live data into Google Sheets in real-time.
As you can see on the right, it maps out the project name and automatically drafts a tailored, context-aware pitch under the “AI Cold Outreach Message” column. Currently testing it at a 15-minute interval, and the token cost for gpt-4o-mini has been incredibly negligible.
Let me know if you guys want me to clean up the .json blueprint and drop it here!
Here it is! I’ve cleaned it up and packaged it with a full setup guide: 
AI Auto-Pitch Radar: Freelancer & LinkedIn Dual-Engine Blueprint
It’s $9 — includes the blueprint + PDF setup guide. Happy to answer any questions!