I hope this will come later with agentic LLMs. Currently the LLMs are getting better at iterative continual conversations, which is a good step.

→ View original post on X — @marek_rosa