nice write up from the HuggingFace folks aggregating works on defining agents, harnesses, environments, RL, etc. The more we can roughly have a shared vocabulary the better…I still find it confusing (lolll), but we’re roughly converging on what everyone is doing
At LangChain we’ve been trying to keep a clean-ish definition and share with the community for agent building pieces, a bunch of the linked works like Addy’s are great
I also find Prime Intellect’s blogs/docs to be excellent at showing where envs, harnesses, and agents fit together
also highly recommend reading HF’s guide on “The Ultimate Guide to RL Environments” linked in article