AutoML for Embedded, developed by Analog Devices (ADI) and Antmicro, is an open-source plugin for Visual Studio Code that works alongside ADI’s CodeFusion Studio plugin. Built on the Kenning framework, it automates the full ML pipeline, including model search, hyperparameter tuning, optimization, compression, and deployment, making edge AI development easy on resource-constrained devices.
The company mentions that it supports the ADI MAX78002 AI accelerator MCU, the MAX32690 MCU, Renode-based simulation, and Zephyr RTOS. It uses SMAC and Hyperband algorithms for automated model search and hyperparameter tuning, along with model compression and quantization to meet strict memory and compute limits. The plugin offers built-in benchmarking for inference speed, memory footprint, and accuracy, while performing RAM and compute compatibility checks. All these features make it useful for applications like image classification, anomaly detection, predictive maintenance, NLP, and action recognition on low-power IoT and embedded systems.
AutoML Embedded overview:
- Type – Open-source AutoML plugin for VS Code that integrates with the CodeFusion Studio
- Framework – Built on the Kenning framework for hardware-agnostic AI optimization and deployment
- Supported Hardware
- ADI MAX78002 AI Accelerator MCU
- ADI MAX32690 MCU
- Extendable to other embedded devices via custom Kenning pipelines
- AutoML Features
- Automated model search and hyperparameter tuning using SMAC and Hyperband
- Model compression and quantization to fit strict memory and compute limits
- RAM and compute compatibility checks before deployment
- Benchmarking Tools
- Inference speed (latency)
- Memory footprint
- Accuracy evaluation
- Real-time performance monitoring
According to ADI and Antmicro, AutoML for Embedded is built as a hardware-agnostic tool based on the Kenning framework, meaning it can work with other microcontrollers and AI accelerators other than MAX78002 and MAX32690. But, the vendor-provided optimizations, like as the AI8X runtime for MAX78002 and microTVM for MAX32690, are specifically designed for those two platforms. For other MCUs, you may need to define your target profiles and pipelines using Kenning’s existing modular architecture.
The company also has different development tools available for MAX78002 and MAX32690 MCUs. For MAX32690, there are a total of four evaluation kits available, where the AD-APARD32690-SL, an Arduino form-factor board based on the MAX32690 ARM Cortex-M4 MCU, and the MAX32690EVKIT are for in-depth evaluation of the same MCU. For battery management, the EVAL-ESS1-SYS provides a scalable BMS platform for cell and pack monitoring, while the EVAL-ADIN6310T1LEBZ serves as a reference design for ADIN6310 field switch applications.
For the MAX78002 MCU, the MAX78002EVKIT comes with 8MB QSPI SRAM, a touch-enabled 2.4-inch TFT display, a stereo audio codec, an I2S microphone, and microSD storage. It supports cameras, audio peripherals, and analog sensors, with all GPIOs accessible via 2.54mm pitch headers. A built-in power accumulator tracks energy consumption for industrial automation, smart cameras, and portable medical devices.
Back in 2020–2021, we saw AutoML used in devices like Google Coral Dev Board Mini, ModBerry AI gateway, and Imago VisionAI cameras using AutoML Vision Edge for on-device model training and deployment. Now, in 2025, AutoML enables support for resource-constrained embedded devices with tools like Analog Devices’ AutoML for Embedded plugin.
The AutoML for Embedded plugin can be directly downloaded from the VS Marketplace, and the source code is available on GitHub. More information is also available on the press release.
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.
