Hi everybody, I had been stucked using other llms that didn't have memory by default or N8N platform, I found this workaround that can be helpful. The first thing that you need to have in mind is that create a thread id in the website that is send and returned in the reply. And save the Id in an html element that will be used for the rest of the season. Second you need to create a database in make.com You'll check the id if exists and is send to the webhook and saved, perhaps is a silly thing but I have ben struggled with this sometime. Finally each x hours you remove "old" conversations and thats all. Hope it helps <div style="position: relative; padding-bottom: 56.25%; height: 0;"><iframe src="https://www.loom.com/embed/96978b7195064a358c521465b9bdd2ff?sid=7f6be311-dc1f-4b17-aa5e-77227ee2961f" frameborder="0" webkitallowfullscreen mozallowfullscreen allowfullscreen style="position: absolute; top: 0; left: 0; width: 100%; height: 100%;"></iframe></div>

Hi there, Anto here, exlibrarian here, I'm over the last year from my library work to a energy company, thans to start in the automation word. Now with impostor syndrome, but many people amazed by my skills. Hope to learn and start a new business and can apply whatever I learn from AI. Thanks everybody