EmbodiedClaw: Conversational Workflow Execution for Embodied AI Development

2026-04-15

AI Summary

This work addresses the substantial engineering overhead and fragmented workflows in embodied AI research under multi-task, multi-scenario, and multi-model settings, particularly in environment construction, trajectory collection, and model evaluation. To tackle these challenges, the authors propose a dialogue-driven automated workflow paradigm that abstracts high-frequency, high-cost development activities into executable skills and leverages natural language instructions to automatically plan and execute end-to-end development pipelines. The proposed system integrates modules for environment creation and revision, benchmark conversion, trajectory synthesis, model evaluation, and asset expansion, thereby significantly reducing manual intervention and enhancing the executability, consistency, and reproducibility of embodied AI workflows.

Technology Category

• Intelligent Robots: Embodied AI
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Embodied AI research is increasingly moving beyond single-task, single-environment policy learning toward multi-task, multi-scene, and multi-model settings. This shift substantially increases the engineering overhead and development time required for stages such as evaluation environment construction, trajectory collection, model training, and evaluation. To address this challenge, we propose a new paradigm for embodied AI development in which users express goals and constraints through conversation, and the system automatically plans and executes the development workflow. We instantiate this paradigm with EmbodiedClaw, a conversational agent that turns high-frequency, high-cost embodied research activities, including environment creation and revision, benchmark transformation, trajectory synthesis, model evaluation, and asset expansion, into executable skills. Experiments on end-to-end workflow tasks, capability-specific evaluations, human researcher studies, and ablations show that EmbodiedClaw reduces manual engineering effort while improving executability, consistency, and reproducibility. These results suggest a shift from manual toolchains to conversationally executable workflows for embodied AI development.

Problem

• Embodied AI
• workflow automation
• development overhead
• conversational interface
• multi-task learning

Innovation

• Conversational AI
• Embodied AI
• Workflow Automation
• Development Toolchain
• Executable Skills