When it comes to creating an AI agent there are many agentic frameworks out there. That makes it easy for you to build an AI agent. And you get to switch between them based on the use case.
I always tell anyone who is starting into the agentic development to start with either Langchain or the Google ADK. I happen to do it both at the same time based on the use case.
Take example of the case where I wanted to create a simple agent. So I decided to use Google ADK. And I documented the process of same in this video.
Google ADK has multiple SDKs in different language bindings. Like Python, Typescript, Golang and now with the Java and kotlin too. I am sure in near future there would be .net support too. I happen to stick with the Typescript and the Python.
Why typescript and Python? Lot of training by the coding agents on these two language bindings. Especially the Python happens to have a lot of agents built with it. And then comes the typescript which is a good binding language to use for most of the web based projects.
If you happen to use the agents for the Saas development, I would prefer the python or the Typescript. For me it turned out to be a good direction with the typescript for most of my web based apps.
Which language you use is upto you. For me what works is these two language SDKs. Have you created your first AI agent with the Google ADK? What was your experience?
Top comments (0)