Samuel Covin On AI Abstractions
If you’re running a customer service business and you have loads of people sitting answering telephones, the less well trained they are, the less that you trust them, the more that you need to give them a script to go through. […] If you’re doing high net worth banking, you just employ people who you think are going to be charming to other rich people and set them off to go and have coffee with people. […] And the same is true of models. The more intelligent they are, the less we need to tell them, like structure what they go and do and constrain the routes in which they take.
If models are getting faster as quickly as you say they are, then we don’t need agents and we don’t really need any of these abstraction layers. We can just give our model […] access to the internet, cross our fingers and hope for the best. Agents, agent frameworks, graphs, all of this stuff is basically making up for the fact that right now the models are not that clever.
Samuel Covin [~00:26:32]
One of many great tidbits from Samuel in this podcast.
I am generally not a fan of Python (in favor of the clearly far superior Javascript) but I am a superfan of Pydantic. I was thrilled when Pydantic AI was announced and have continued to follow it’s developments and iterations closely. I have a strong feeling it will continue to define mental models in the AI SDK space for many years to come.