Santiago (@svpino) says connecting a coding agent, such as the Codex CLI, lets it control a simulated robot by writing and running Python code.
The agent checks results and saves reusable procedures as skills, which can transfer to a real robot in a different environment without retraining a model.
The post links to EmbodiedEvo, a project from Einsia AI that reports a 45.08 average score and 38.95% success and ranks first on the RoboDojo leaderboard.
If you connect this to your coding agent, you can control a simulated robot and build experiences you can reuse in other scenes.
They have an example using the Codex CLI.
You can then reuse those lessons with a real robot in a different environment. No need to retrain a model.
Coding agents are entering the physical world. Zero-shot is only the starting point. What if we gave them practice—and let the next agent inherit what they learned? EmbodiedEvo: one Playground → reusable Skills & Lessons → unseen layouts. No model-weight updates. 45.08 average score. 38.95% success. #1 on the RoboDojo leaderboard. Project: https://lab.einsia.ai/embodiedevo/ Paper: https://arxiv.org/abs/2610.18867 Github：https://github.com/Einsia/EmbodiedEvo
Source: Santiago · x.comPublished
