Embedr is an AI-powered Arduino-like IDE designed to simplify development for Arduino-compatible microcontrollers. It looks very similar to the Cursor code editor and uses Microsoft’s Monaco Editor and Arduino CLI tools to enable features such as intelligent code suggestions, project skeleton generation, and visual debugging.
Embedr has all the Arduino features that you are familiar with, and on top of that, you can add other LLMs and experiment just by adding the required plugin and connecting them via API keys. We’ve previously seen Arduino release their Claude-based Arduino Cloud Editor, which runs in the browser for context-aware coding assistance. But the problem with that is, if you don’t have internet, you can’t use the IDE. Embedr runs locally and uses Google’s Gemini 2.5 Flash LLM, meaning you can compile and debug your code offline without needing a constant internet connection. Additionally, Embedr gives you greater control and privacy compared to the cloud-only, single-model setup of the Arduino Cloud Editor.
Main features of Embedr AI IDE
- Platform – Runs locally on your computer (not browser-based)
- Compatibility – Works with Arduino-compatible boards via Arduino CLI
- AI Model – Powered by Google Gemini 2.5 Flash LLM
- Editor – Built on Microsoft Monaco Editor (used in VS Code)
- Main features
- Natural language to Arduino sketch generation
- Inline code completion and AI-assisted debugging
- Project skeleton creation using prompts
- Visual debugging interface and tools
- Works offline with full compile and debug support
- Supports external LLMs via plugin and API key
- Use Cases
- Debugging code with AI
- Rapid development with LLM-generated code
- Exploring multiple AI models in Arduino workflows
- Learning Arduino with intelligent code suggestions
When I first found out about this, I wanted to try it immediately, ’cause I was vibe coding with Gemini for a couple of months now, and so far it has given me very decent results, to say the least. I was able to code a complete ESP32-based home automation dashboard in like 30 minutes or so, which included buttons for appliances, an RGB LED controller, a fan speed controller, and a section for showing temperature and humidity data – all with manual controls for switches and fan controllers.
I installed the IDE, opened it, and was immediately disappointed to see that there’s nothing for ESP32. But anyway, I selected Arduino Uno WiFi and used the example prompt for a home security system. It first generated a suitable name for the project, then defined the system design. Next, it generated the code and included the WiFiLink library, but the compilation failed. It tried to install the library but failed again. Then it removed the library, modified the sketch, and the code finally compiled. All in all, the tool looks good, but the lack of support for ESP32 is disappointing.
The project is in early development, and you can find more information about it on its official website. The IDE is free to use for up to 20 requests, 500 completions/month. After that, you have to pay $3/month for 800 requests, and $24/month for unlimited use. The source code is on GitHub under a non-commercial license, which may conflict with the MIT and GPLv3 licenses used by Monaco Editor and Arduino CLI.
Via Hackster.io
Debashis Das is a technical content writer and embedded engineer with over five years of experience in the industry. With expertise in Embedded C, PCB Design, and SEO optimization, he effectively blends difficult technical topics with clear communication
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress. We also use affiliate links in articles to earn commissions if you make a purchase after clicking on those links.
