Atmosis swaps raw data for intelligible reports

Atmosis turns environmental data into easy-to-understand reports. The device uses edge AI and integrates Gemini LLM to contextualize information. COP31, in Antalya, will be the stage for the presentation.

The Atmosis device, developed by Next Builder, turns environmental data into easy-to-understand reports. It uses the Arduino UNO Q and sensors such as DPS310, X6 and a dust sensor to collect information on temperature, pressure, humidity and fine particles. The data is analyzed with edge AI via Edge Impulse, which alerts users to conditions that require attention.

Atmosis swaps raw data for intelligible reports — Image: blog.arduino.cc

In addition, Atmosis can integrate Gemini LLM to offer personalized answers to questions about the collected data. The results are displayed via web and Telegram, making access and interpretation easier. COP31, the United Nations Climate Change Conference, begins on November 9, 2026, in Antalya, Turkey. The conference seeks to strengthen cooperation on solutions that reconcile environmental conservation and social well-being. Climate-Informed Engineering (CIE) promotes the integration of accurate climate data into engineering practices, something Atmosis supports with its contextualized approach. Major environmental groups are adopting AI to analyze climate data, track wildfires and more, even as they warn about the environmental impact of data centers.

Atmosis is an initiative that aims to simplify the interpretation of environmental data, with a focus on PM2.5 and other indicators. Atmosis uses the Arduino UNO Q as its processor, which allows fast and efficient operations with the integrated sensors. The DPS310 sensor measures pressure and temperature accurately, while the X6 monitors humidity and luminosity. Together, they offer a complete view of the environment, without requiring additional equipment.

The dust sensor detects particles in the air, a useful feature in areas with high pollution. The LED strip serves as a visual indicator, quickly showing whether the data is within the expected limits. The compact 3D-printed enclosure ensures the device is light and portable, making it easy to install in a variety of places. The integration with Gemini LLM allows Atmosis to answer specific questions, making the reports more interactive and personalized. This feature is especially valuable for engineers and managers looking for direct solutions to environmental problems. The dependence on the internet for access via web and Telegram may limit use in areas with poor connectivity, but the integration with edge AI reduces the need for a constant connection.

The initiative, developed by Next Builder, aims to simplify the interpretation of indicators such as PM2.5, temperature and humidity by offering contextualized information. The technology is based on edge AI and integration with Gemini LLM, enabling personalized responses and dynamic reports. The device, which fits in a compact space, uses the Arduino UNO Q as its processor, ensuring fast operations with the DPS310, X6 and dust sensor sensors. For engineers and managers, Atmosis can be a valuable tool, but its price and availability in Brazil have yet to be confirmed.

The device generates easy-to-understand reports using edge AI and Gemini LLM. This helps engineers and managers better understand local conditions. The technology makes it possible to collect data accurately, thanks to the Arduino UNO Q and the DPS310, X6 and dust sensor sensors. The system displays information via web and Telegram, making real-time access easier. The LLM integration offers personalized responses, turning data into concrete actions.

With information from blog.arduino.cc, exame.com and forbes.com.