Hey everyone,<p>It's Mart, Neel and Jack from the Factorio Learning Environment team.<p>Since our initial release, we have been working hard to expand the environment to support multi-agent scenarios, reasoning models and MCP for human-in-the-loop evals.<p>We have also spent time experimenting with different ways to elicit more performance out of agents in the game, namely tools for vision and reflection.<p>Today, we are proud to release v0.2.0, which includes several exciting new features and improvements.<p>Thanks for checking this out.
I have a few qualms with this app:<p>1. For a competent ML researcher, this is trivial to build using VQA + MCTS + standard message passing.<p>2. MARL techniques with attention-based communication solved this years ago<p>3. It does not seem very "viral" or income-generating. I know this is premature at this point, but without charging users for the service, is it reasonable to expect to make money off of this?