GuideBB-0206

Using AI agents in VS Code to build, test and debug STM32 firmware

Tech-AD-venture's 23-minute video walks through using AI agents in VS Code for STM32 firmware, contrasting autocomplete with task-based agents. Tools covered include GitHub Copilot, Claude, Codex and PlatformIO. An STM32 RTOS example runs from requirement to review. Its description lists MCP hardware access as a possibility; we could not confirm on-board debugging in the video.

Open original source youtube.com · Tech-AD-venture

Claimed
Why this evidence label?

Description lists the workflows; we did not confirm on-board debugging in the video.

How we check entries
Play videoUsing AI agents in VS Code to build, test and debug STM32 firmwareThe player loads from youtube-nocookie.com when you press play.
The source video, on YouTube.

Briefing

The video follows an earlier one on the embedded development lifecycle. The STM32 example runs requirement, implementation, build, test, debug and review in order. Debugging covers compiler and linker errors, call stacks, registers and HardFault analysis, plus UART log analysis. Besides the agents, the description names clangd, CMake, Cortex-Debug, a serial monitor and Git.

The description treats Git feature branches, diffs, tests and engineer approval as steps that stay with the engineer. It says only that AI can potentially connect through MCP to diagnostics, sensor data and controlled hardware interfaces. It names no board beyond STM32 and no debug probe.

Tags

Board
Interface

Report a problem with this entry