things to try: - slow chat - with tools - streaming chat - with tools video ideas: - [ai agents elixir conf](https://www.youtube.com/watch?v=Wu3jXi1IyeM) - recommends oban instead of GenServer - can autoscale with flame - tool calling scaling? - using elixir langchain for tool definitions real time blog: elixir clustering examples and architecture: process groups for distributing workers across cluster: zoi validation library: