Skip to main content
The agent is only as strong as the expertise behind it, and that expertise comes from engineers, not the model. The benchmark gap is the proof: same model on both sides, won by curated firmware knowledge and a clean architecture. The agent gets stronger as that knowledge base grows, and there are two ways to get involved.

​
Contribute knowledge (embedded experts and specialists)

This is the part that makes the agent good: the hard-won fixes and idioms you only get from shipping nRF and ESP firmware. We are building a dedicated studio for authoring this expertise and will open it to outside specialists once it has earned its keep in-house. What we are looking for
  • The runtime failure modes that build clean and pass review, then fail on hardware.
  • Board, SoC, and Kconfig traps: the configuration that is technically valid but wrong for the part.
  • Protocol knowledge: the BLE, Wi-Fi, and connectivity behavior that only makes sense once you have debugged it on real silicon.
What you get
  • Credit for your work.
  • A founding contributor’s say in how the studio takes shape.
  • You keep the rights to what you contribute.
  • Early access to the authoring studio as it comes online.
If you have lived inside these failure modes, start a discussion.

​
Contribute code (open-source developers)

The extension is open source (Apache-2.0, built on Cline). Improve the tool itself, or add a benchmark task in evals/ (each task ships its bug-injection patch, reproduction procedure, and known-correct fix). Open an issue or PR.