Skilled AI Agents for Embedded and IoT Systems Development
This addresses the problem of automating software development for embedded and IoT systems, which is incremental as it builds on existing agentic systems by adding hardware-specific skills.
The paper tackled the challenge of applying AI agents to hardware-in-the-loop embedded and IoT systems development, where code often fails on real devices due to hardware-specific issues, and introduced a skills-based framework with a benchmark showing that human-expert skills achieved near-perfect success rates across platforms.
Large language models (LLMs) and agentic systems have shown promise for automated software development, but applying them to hardware-in-the-loop (HIL) embedded and Internet-of-Things (IoT) systems remains challenging due to the tight coupling between software logic and physical hardware behavior. Code that compiles successfully may still fail when deployed on real devices because of timing constraints, peripheral initialization requirements, or hardware-specific behaviors. To address this challenge, we introduce a skills-based agentic framework for HIL embedded development together with IoT-SkillsBench, a benchmark designed to systematically evaluate AI agents in real embedded programming environments. IoT-SkillsBench spans three representative embedded platforms, 23 peripherals, and 42 tasks across three difficulty levels, where each task is evaluated under three agent configurations (no-skills, LLM-generated skills, and human-expert skills) and validated through real hardware execution. Across 378 hardware validated experiments, we show that concise human-expert skills with structured expert knowledge enable near-perfect success rates across platforms.