Simantic is the simulation layer that lets engineers and AI agents test and validate firmware at the speed of software.
Frontier AI models have made generating firmware code cheap and effortless, but testing still depends on manual hardware debugging, expensive equipment, and software designed more than a decade ago. Simantic is modernizing firmware testing with simulation built for the era of coding agents.
w/ @ycombinator@shong0504@enough_ahnaf
Ever programmed a satellite before?
Well I have been for the last month and its all been inside simulation in @simantic . The internals of consumer hardware are not so different from space tech.
They did send apollo to the moon using a potato chip of a computer.
Can an ESP32-S3 run AI models locally without sending data to the cloud?
That's the idea behind esp32-s3-edge-ai from Simantic Dev.
The project explores running edge AI on the ESP32-S3, a dual-core microcontroller clocked at up to 240 MHz, with vector instructions for accelerating neural-network operations.
The interesting challenge is fitting inference into limited RAM and flash while keeping latency and power consumption under control.
For anyone working with embedded AI, this is a useful direction to explore.
Think about local sensor classification, smart devices and always-on detection without depending on a cloud connection.
I would be particularly interested in seeing how far INT8 quantisation, PSRAM and the ESP32-S3 vector instructions can push on-device inference.
Would you try running your own trained model on an ESP32-S3?
ESP32-S3 Facial recognition running in Simulation!
Edge AI and IOT is on the rise and we are here to help it get there.
This is a Virtual ESP32-s3 running real firmware connected to the video feed of my webcam via serial.
Of course it's insanely laggy but thats mostly because demoing using an inefficient path, you would be using the esp32s3's camera interface directly. But even then you would probably get 3 FPS, bargain deal when you know that an s3's cost like 6 bucks.
Long time fan of the ESP32 series and so excited to keep sharing with everyone all the things you can do in simulation!
It's gonna be a week of launches! But if you want to beat me to making cool things run in simulation, try it -> simantic.com
ESP32-S3 Facial recognition running in Simulation!
Edge AI and IOT is on the rise and we are here to help it get there.
This is a Virtual ESP32-s3 running real firmware connected to the video feed of my webcam via serial.
Of course it's insanely laggy but thats mostly because demoing using an inefficient path, you would be using the esp32s3's camera interface directly. But even then you would probably get 3 FPS, bargain deal when you know that an s3's cost like 6 bucks.
Long time fan of the ESP32 series and so excited to keep sharing with everyone all the things you can do in simulation!
It's gonna be a week of launches! But if you want to beat me to making cool things run in simulation, try it -> simantic.com
What if you could test embedded firmware without having the actual development board connected?
Peripheral Pretender by Clifford Swartz takes this idea quite far.
It creates a QEMU-based digital twin of Microchip development boards and MikroE Click modules, allowing you to run your firmware in a virtual hardware environment.
You can load your compiled ELF, interact with virtual buttons and sensors, monitor UART, SPI and I2C traffic and inspect GPIO activity through a logic-analyzer-style timeline.
There are some particularly useful debugging features. You can inject hardware faults, such as shorting a pin or forcing an I2C device to return NACK, then see how your firmware handles them.
It also supports deterministic recording and replay, so you can reproduce a test and identify exactly where behaviour starts to differ.
This could make firmware development, regression testing and hardware debugging much easier.
Would you use a digital twin like this in your firmware testing workflow?
Today we released Pyrite 0.10.0. Pyrite is the most accurate and complete firmware simulation model for Arm Cortex-M/R and RISC-V based platforms, made by @simantic.
Pyrite is out of stealth, and the usage documentation is available at simantic.com/docs
A lot of people don't know this, but you don't need a GPU, let alone an operating system, to run a neural network. We just ran one on a $20 single-core Arm Cortex-M7 microcontroller programmed entirely on bare-metal C.
We ran a neural network on a simulated model of the STM32H753 microcontroller using the Pyrite sim from @simantic. Here's a quick demo:
BLE is a vital part of IOT in the world and really nice to see more and more efforts to make it modeled and safe for development.
Too many times there are closed source blobs(looking at you Espressif) and you hope and pray there isnt vulnerabilities.
We are doing similiar train of thought research in our autonomous lab in @simantic to enable firmware development that is safe and reliable with simulation.
arxiv.org/html/2608.29737v1