Hi everyone, excited to join the community.
At Hashlogics, we’ve been actively building AI powered retention and loyalty systems using Zapier as the core automation layer. Our current focus is around agentic workflows where AI agents trigger personalized actions across tools like Slack, Notion, and email, based on live user behavior.
One scenario we’re testing:
-
A user event enters via webhook
-
Zapier calls a GPT based agent for segmentation and intent detection
-
It then routes the response to dynamic follow up actions through Slack or scheduled sequences
-
All results are logged in Notion for async visibility
We’re deliberately keeping the setup lean and dev friendly so it’s scalable without depending on a complex custom backend. I’ve personally blocked time to handle this sprint and would love to connect with anyone solving similar challenges, especially around:
-
Structuring testable Zapier workflows for AI agents
-
Handling unpredictability in LLM outputs
-
Keeping tone and logic consistent across multi step automations
If it helps, happy to share more of our process or compare notes. Looking forward to learning from others here.