Series A Startup Dryad Networks leverages Omdena Top Talent AI Teams to Advance Early Wildfire Detection

Dryad Networks is a young German company with a clear mission: stop wildfires before they spread and help fight climate change through ultra-early detection. The founders believe prevention is more effective than reaction, and early sensing is essential. In this partner interview, they explain how collaboration with Omdena’s Top Talent AI teams is accelerating their technology and how recent Series A funding will help scale their solution. Their journey shows how investment, field testing, and machine learning work together to advance wildfire detection.

As climate change intensifies droughts and heat waves, wildfires are becoming more frequent and destructive. Traditional methods such as towers, cameras, and satellites usually detect fires only after flames or smoke appear — often too late for easy containment. Dryad’s approach focuses on the earliest possible warning. Its solar-powered sensors detect changes in gas at the forest floor, identifying fires at the smouldering stage when a spark first ignites organic material. Detecting these micro-events cuts response time and helps prevent large-scale damage, shaping every strategic decision the company makes.

Scaling up with Series A funding

The most recent success for Dryad is securing €10.5m in Series A Funding to bring ultra‑early wildfire detection to the global market. Fundraising is not just about money; it signals market confidence and provides the resources to transform a promising prototype into a deployable product. The round was led by eCAPITAL and included additional investors such as Toba Capital, Semtech and TIME Ventures, the investment fund of Marc Benioff. By participating in the round these investors showed faith in Dryad’s vision and technology. The capital infusion enables the company to scale its team, accelerate its go‑to‑market strategy and fulfil its mission to fight climate change and protect forests around the world.

For a hardware‑intensive start‑up like Dryad, scaling up means more than just hiring engineers. It involves establishing manufacturing partnerships, building supply chains, expanding field operations and developing software and machine‑learning capabilities in tandem with the physical product. Series A funding provides the runway to hire specialists in these areas, secure components at scale and maintain inventory, and invest in research and development to refine both sensors and algorithms. Without this support, even the most promising environmental innovations can stall. The funding thus represents a critical milestone on the path from prototype to global impact.

Testing and deploying ultra‑early wildfire detection technology

Dryad is validating its system through ten proof-of-concept trials across southern Europe, the United States, and Asia. These tests expose sensors to a wide range of climates and terrains, from dry Mediterranean forests and windy conditions to damp Pacific Northwest landscapes and Asian monsoon regions. This diversity ensures the system performs reliably in real-world conditions.

As Dryad moves beyond pilots into broader deployment, medium-scale rollouts are planned for later this year. Its manufacturing partner Prüfrex will produce 10,000 sensors in October and 230,000 units in 2023, applying strict quality control so Dryad can focus on design. These sensors power the Silvanet network, capturing gas data from the forest floor to detect fires at their earliest smouldering stage.

Alongside production, Dryad conducts controlled burns across Europe to train its sensors and machine-learning models. Engineers expose devices to heat and smoke to analyse how gas levels change and distinguish normal environmental shifts from combustion. After a burn in Nuremberg, sensors were left to monitor lingering gases from dying embers, refining detection accuracy, reducing false alarms, and improving system reliability.

Source: Dryad

The first image shows a Dryad team member holding a green sensor in a forest clearing while smoke from a prescribed burn rises behind him. The photo highlights the small size and easy deployment of the device, while the burnt ground illustrates the harsh conditions the technology must withstand. It captures how hands-on field testing and real-time sensor feedback are central to Dryad’s mission of ultra-early wildfire detection.

During an exhibition at INTERFORST in Munich with Bosch Security and Safety Systems, Dryad unexpectedly received a sensor alert. At first, the team suspected a false alarm, but they contacted fire services as a precaution. The alert proved real: a zombie fire had survived the prescribed burn and reignited due to shifting winds. These fires smoulder underground for long periods without visible flames and can flare up suddenly. This became the first unplanned wildfire detected by Silvanet, and because it was caught early, firefighters quickly contained it before it could spread.

Smoke drifting through the forest from a hidden smouldering fire

The second photograph captures a stand of tall, slender trees bathed in light, with smoke drifting through the trunks. It evokes the eerie beauty of a forest when a fire is smouldering but not yet out of control. Thin columns of smoke provide an early visual clue, but the delay between smoke appearance and detection through human observation can be significant in remote areas. Sensors buried at the forest floor, like those Dryad manufactures, respond to changes in gas composition well before such plumes form, giving first responders a head start.

When the alert at INTERFORST came through, Dryad’s team called the local fire brigade immediately. Firefighters arrived quickly, confirmed that embers hidden beneath organic debris had reignited, and doused the site with water. The outcome demonstrates how early detection, swift communication and prompt response can make the difference between a contained incident and a wildfire that consumes hectares. By catching the smouldering fire in time, Silvanet saved resources and prevented air pollution.

Firefighters deploy pumps and gear to suppress the reignited embers

The third image shows firefighters in bright yellow and orange gear operating pumps and hoses in a forest road. Large red water pumps and fire trucks dominate the foreground, while dense trees stand in the background. The scene underscores the collaborative nature of firefighting: technology can provide an early warning, but human crews still need to mobilise equipment, transport water and carry out suppression. The photograph reminds readers that early detection is a tool that supports, rather than replaces, the courage and skill of fire services.

Collaborating with Omdena’s top talent AI teams

Ultra-early detection depends not only on sensors but also on advanced data analysis. Each unit captures raw gas values along with temperature and humidity, which must be processed to separate meaningful signals from environmental noise. To accelerate this work, Dryad partnered with Omdena through an AI Innovation Challenge, gaining access to a specialised team of 50 AI engineers within weeks.

During the eight-week challenge, the team built multiple data pipelines using sensor data from controlled burns, normal conditions, and the real zombie-fire incident. Because gas readings change with soil type, moisture, and vegetation, engineers applied signal filtering, feature engineering, and supervised learning to detect combustion patterns. They tested models including decision trees, support vector machines, and neural networks, balancing sensitivity and false-alarm control. The final system accurately classifies data as either “in-smoke” or “clean-air” and is scalable across different forests without constant retraining.

The project also delivered an automated pipeline that processes incoming sensor data and generates alerts in real time — a necessity when managing thousands of devices. Dryad continues working with Omdena teams to improve performance through tuning, model ensembling, and data augmentation. New field data is continuously added, enabling over-the-air model updates and ongoing accuracy gains. This closed-loop system — from forest to cloud and back — ensures detection remains reliable as climate and terrain conditions change. These advancements build on a growing ecosystem of AI wildfire solutions, many of which follow similar architectures highlighted in Omdena’s wildfire prevention work.

Key outcomes

• Series A financing: raising €10.5 million led by eCAPITAL with participation from Toba Capital, Semtech and TIME Ventures will fund scaling and global deployment.
• Proof‑of‑concept trials: ten trials across three continents, with medium‑scale roll‑outs and mass manufacturing plans for 2023.
• Field‑derived insights: controlled burnings and post‑fire monitoring provided data on smouldering behaviour, informing sensor firmware and algorithm improvements.
• First real‑world detection: Silvanet sensors detected a zombie fire during a prescribed burning, enabling firefighters to extinguish it quickly.
• Machine‑learning advances: Omdena’s AI Innovation Challenge produced models that classify sensor data into “in‑smoke” and “clean‑air” categories and provided a scalable, replicable pipeline for future improvements.
• Ongoing optimisation: continued collaboration with Omdena top talent teams focuses on optimising training pipelines and integrating feedback from field deployments.

Conclusion 

Dryad Networks’ journey illustrates how innovation, collaboration and investment combine to advance early wildfire detection. The company’s mission resonates in a world grappling with climate change and rising wildfire risks. With fresh capital, a rapidly expanding sensor network and the support of Omdena´s Top Talent AI teams, the start‑up is closer than ever to delivering ultra‑early detection at scale. Its sensors detect changes in gas composition at the forest floor long before smoke becomes visible, and its algorithms turn that data into actionable intelligence for firefighters and forest managers. Each successful proof‑of‑concept trial, every manufactured sensor and each refined machine‑learning model brings Dryad nearer to a future where wildfires are prevented rather than combated.

As wildfires become more frequent and severe, technologies that sense smouldering fires before they ignite are critical tools in the fight against climate change. Early detection reduces suppression costs, protects biodiversity and keeps communities safe. It also buys time for responders to evacuate people, deploy resources and contain fires while they are still manageable. By investing in research, conducting rigorous field tests and embracing collaborative AI development, Dryad sets an example for how start‑ups can tackle global environmental challenges.

If you want to build advanced AI solutions that support early detection, environmental protection and climate resilience, connect with Omdena today. Share your goals and let us see if we are a good fit.