I have created VAPI voice agents for my business which are working well. They have prior call “memory”, can look up property information from another platform, accesses a pinecone database to get our service information and so on. To give the voice agents a persona, I created an avatar in HeyGen. I use perplexity to generate relevant content daily, and use AI to select the most engaging content, create a script, and pass it to HeyGen programmatically via their API. It randomly selects one of the avatar’s looks, and posts the completed video on social media using Blotato. I just want more traction with the posted content. It is geared towards real estate and pest control (my two businesses). Here’s a sample. Feedback would be appreciated.

🔥 Unconscious bias is one of the biggest challenges in recruiting. So I built a workflow to fix that: 1️⃣ Resume comes in through webhook 2️⃣ AI anonymizes it → strips out names, gender, and identifiers 3️⃣ AI scores the candidate based only on skills & experience 4️⃣ Result is saved to Google Sheets 5️⃣ Recruiter gets an email report with the score + explanation ⚡ This ensures evaluations are based on merit, not unconscious bias. 👉 Engagement questions for Skool: Would you trust an AI anonymizer to handle resumes in your pipeline? Do you think this could improve diversity hiring outcomes?

Is anyone having trouble retrieving data from Google Sheets this morning? My Google Sheets node will not retrieve data. I have verified and signed in again with my credentials. I can run a Google Sheet node to delete the oldes 10 rows, but I cannot read from my sheet.

Does anyone else use HeyGen for videos? It has been aweful for me lately. The onsite studio just can’t get it right. And I get an API error when trying to use ElevenLabs Turbo 2.5. Has anyone else experienced that? Even with the other voice models, the voice is terrible no matter what I do. It’s the same voice I use for my voice agent through VAPI and it sounds great there. Is there an equivalent platform to HeyGen? I do also generate videos programmatically via their API and that seems to be working OK, but not great. Is there anything I should be doing?

I have two workflows that work together to generate HeyGen videos. The first seeks out engaging content using perplexity, and saves it to a Google Sheet. My other workflow evaluates the last 5 articles and selects the most engaging one, then generates a script which is used to generate a HeyGen video programmatically via their API. My questions, is it better to keep these workflows in separate workspaces, or place them both in the same workspace in n8n? Does it make a difference? They operate on separate schedules.