Test firmware in simulation before using hardware
Simantic lets agents run firmware against virtual devboards, with a reviewer checking results before any real board is needed.
- What they did
- Simantic gives agents programmable virtual devboards that can be launched from the terminal with a single API call. Behind that is a simulation engine the founders rebuilt in Rust after a Renode-based wrapper turned out to be too slow and too imprecise. A firmware reviewer agent uses the simulation output as evidence for design choices, and CI checks every commit against a simulated microcontroller.
- How it went
- The founders say checks that once took hours on physical devboards now finish in seconds. One early satellite startup used it to try firmware on hardware it did not have, at roughly 1% of the usual cost and time. Support is still limited to certain STM32, ESP32, and RP2040 models.
- Tips
- Confirm your chip is supported before trying to rely on the workflow. This is most useful when you need fast firmware validation without a board on your desk.
Try it yourself with Claude Code
I’m working on firmware for [board or microcontroller] and want to validate it without physical hardware. Set up a simulation workflow for my repo at [repo path], run the firmware in a virtual board environment, and check it against [specific behavior to verify]. Use the simulation results to review likely firmware issues such as timing bugs, register mistakes, or unsafe assumptions. If anything would require a destructive or irreversible action, stop and ask for my approval before proceeding. Finish with a clear pass/fail report and a short list of findings, and tell me whether my chip is supported before you start.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Check traces for hidden agent errors
Laminar’s flow-1 reviews agent traces end to end so quiet mistakes can be caught without sampling.
Build a staged agent pipeline for PRs
Create a multi-step system that assigns issues, tests changes, and leaves the final merge to a person.
Trace C# symbol usage with a Roslyn-backed index
Graphify C# helps agents map real C# usages, including overrides and implementations, from compiler-resolved project data.
Let an AI agent inspect a Windows crash dump
Use MCP to give an AI agent structured crash details from a Windows dump so it can track down the fault and propose a fix.
Turn tickets into code changes and draft the PR
Stripe uses in-house coding agents to carry tasks from ticket to reviewable code, then humans check the result.
Profile and tune a CUDA kernel with an agent
They used a LangGraph agent to test CUDA kernel tweaks, verify results, and profile candidates with Nsight.