From “iron manual” to “Iron Man” — Augmenting GPT for fast editable memory to enable context aware question & answering

TL:DR We show how Marqo can be used as an update-able and domain specific memory to GPT to perform question and answering for products and chat agents. We also show how reference and hallucination checks can be easily implemented. A walk-through with images and animations is provided along with the code to reproduce.