Hey everyone, I would love to understand some of the challenges you face in developing AI voice agents. Based on your feedback, I will create videos to help address these challenges. Thank you!

Hey everyone! I would love to get some advice. Basically, I have been almost unable to get a single client for a while. Fortunately, I am finally mostly about to convert my first client. However, I am really struggling with making the proposal. Basically, the main problem I'm facing is the following: They want me to make an internal chatbot for a project they are doing. They have sensitive data. Now the main issue is, this data seems to be from old PDFs that they are currently in the process of digitizing, and then I would need to make a Vector DB from this. However, the size of the data is in the multiple TeraBytes (1000 GB = 1 TB) and im assuming it would be around 5 TB of digitized pdfs that i would need to convert into a db, now after doing some searching a lot of sources have said an approx cost would be 60000USD to do it for 5 TB but the total charge for the project seems to be around 8000USD. If that's the case, I would unfortunately be forced to drop the project. Basically, the main point of concern is not the automation but the data pipeline itself. How would I go about this? Please, any advice would help I've been wracking my brain for a day. To summarise since im panic rambling: 5TB digitized pdfs present that i need to convert into a structured vector db for a rag chatbot (the rag pipline will be with n8n of course) but unsure about the processing costs to make the db and what the monthly db costs would be

Hello. I overheard someone say during the most recent call that voice agents built on the GHL platform are not performing well. Can't remember who said it. But I would like to get some details about it. I'm not building using GHL. Just would like to know more so I can also inform people I am talking to who are asking about using it. Thanks!