Which AI chatbot do you use to brainstorm and improve your RL training?

Hey,

I’m working on a RL project with a coach/trainer module, and I regularly brainstorm with AI chatbots (Claude, ChatGPT, Gemini) to analyze decision quality, debug training issues, and find improvements.

The problem: this back-and-forth is very time-consuming, and I’m looking to optimize it.

A few questions:

1. Which chatbot do you find most effective for RL-specific brainstorming (policy issues, reward design, training instabilities…)? 2. Any prompting strategies or workflows that save you time? 

Looking for feedback from people who’ve used LLMs seriously on real RL projects. Thanks!