Reimagining Wildfire Intelligence from Space
When wildfires tore through Southern California in the spring of 2025, entire towns were enveloped in smoke within hours. Aerial firefighting aircraft were grounded due to low visibility, forcing delays in containment efforts. An unauthorized drone entered restricted airspace and collided with a suppression plane, halting operations entirely. Across the western U.S., emergency response teams faced yet another fast-moving disaster without the kind of real-time, coordinated intelligence needed to act with speed and precision. This isn’t just a failure of equipment, will, or effort. It’s a failure of infrastructure.
For centuries, the default strategy for wildfires has been rudimentary: find the fire, put it out. But in a world of rising temperatures and longer dry seasons, that playbook has reached its limits. Fire is a natural part of many ecosystems, and some regions today actually suffer from a lack of regular, lower-intensity burns. Without them, vegetation builds into dangerous fuel loads that, once ignited, erupt into the kinds of megafires that now dominate the headlines. Essentially, we don’t have too much fire—we have too little of the right kind.
Addressing this imbalance requires a complete rethinking of how we monitor, understand, and manage wildfire events globally. That’s the mission behind FireSat, a dedicated satellite constellation led by Muon Space in collaboration with the Earth Fire Alliance, a nonprofit coalition formed to shift wildfire response from emergency reaction to long-term ecological resilience. The Alliance convenes over 200 experts from fire science, forestry, and emergency management to help reframe how we relate to fire at a planetary scale. Google is another foundational partner, with Google Research playing a prominent role in scaling the platform’s impact. At Google I/O 2025, Google CEO Sundar Pichai highlighted Muon Space and FireSat as a cornerstone of its wildfire resilience initiative, building on years of Google’s product-level work that includes wildfire boundary detection in Search and Maps, and the release of its synthetic Firebench dataset on Google Cloud to support machine learning research in fire propagation. The effort is further backed by the Gordon and Betty Moore Foundation, a longtime leader in wildfire policy and land conservation, and the Environmental Defense Fund (EDF), which brings decades of experience in environmental monitoring and climate advocacy.
At the heart of FireSat is a belief that truly comprehensive data, when delivered faster, and built for purpose, can unlock a more strategic approach to wildfire management. Designed to provide high-fidelity insights across every stage of a fire’s lifecycle, FireSat aims to equip frontline responders, scientists, and policymakers with the kind of global fire intelligence infrastructure that has long been missing. We discussed this in more detail with a panel of experts at our 2025 Space Capital Summit. At full capacity, the FireSat constellation will consist of more than 50 satellites, each equipped with advanced six-band infrared sensors tuned specifically for wildfire detection and state of the art RF capabilities. The system revisits every point on Earth every 20 minutes—and wildfire zones every nine—tracking fires in near real-time. It captures movement, intensity, and growth with five-meter resolution across all terrains.
This level of fidelity enables a fundamental shift in strategy. Even as the tools of modern wildfire response continue to advance—autonomous drones scouting burn zones, AI models forecasting spread, robotic firebreak systems, smart tankers, and IoT-enabled ground sensors—none of them function effectively in isolation. Without a trusted, high-cadence intelligence layer to coordinate where, when, and how these assets are deployed, response remains fragmented and delayed. Just as crucial, fires in low-risk areas should be allowed to burn under controlled conditions as part of broader ecological stewardship, rather than being extinguished by default.
FireSat fills all this gap. It acts as both the overwatch and connective tissue for a new class of wildfire operations: distributed, intelligent, and globally scalable. Its AI-driven analytics engine continuously compares incoming satellite imagery against thousands of historical observations, integrating real-time weather and terrain data to flag emerging ignition points. With clear visibility into which fires are spreading rapidly, burning intensely, or threatening critical infrastructure, response teams can finally prioritize their actions with greater precision. It marks a transition from reactive suppression to proactive, data-informed fire management—supporting not only emergency situations, but also enabling researchers to model fire behavior, land managers to plan prescribed burns, and policymakers to shape long-term climate resilience strategies.
The Right Team for the Hardest Missions
FireSat is, by any measure, a leap forward in reframing how we observe and manage one of the planet's most complex environmental threats. Crafting such a sophisticated system is only possible with a team possessing with the rare blend of vision, technical depth, and executional excellence, for they have not only built and scaled best-in-class satellite systems before but inherently understand what it takes to bring ambitious ideas to life.
At the helm of Muon Space is CEO and Co-Founder Jonny Dyer, whose career spans some of the most consequential Earth observation (EO) milestones of the past two decades. As Chief Engineer at Skybox Imaging, Jonny was part of the Stanford dorm-room team that set out to challenge one of the most entrenched assumptions in space infrastructure, which is that high-quality imaging from orbit requires a billion-dollar spacecraft. Skybox proved otherwise.
In the early 2010s, the EO industry was still dominated by a small handful of legacy providers whose large, exquisite satellites operated on decade-long timelines and Cold War-era budgets. Skybox’s ambition wasn’t just to build smaller satellites, it was to make space data faster, cheaper, and more useful. That meant applying commercial-grade camera design, software-defined systems, and consumer electronics principles to the problem of orbital imaging. Their first satellite, SkySat-1, launched in November 2013, was the first smallest satellite in history to deliver sub-meter, high-resolution video from space, capturing 90-second clips at 30 frames per second from a platform that could be built and launched in months, not years. Skybox’s success culminated in a $500M acquisition by Google in 2014, forming the basis for its satellite imagery division.. Skybox’s fast-iterate, full-stack philosophy made orbit and made history. Even Planet Labs, whose early Dove satellites faced challenges, eventually acquired the SkySats from Google to augment its capabilities with higher-resolution data.
But for Jonny, the experience revealed a deeper insight: building a technically excellent satellite wasn’t enough if users couldn’t readily interpret or act on the data. Skybox had proven what was technically possible, but integrating that data into real-world decision-making remained a downstream challenge. In EO, the hard part isn’t just sensing, it’s turning that sensing into timely, meaningful insight. The realization stayed with him, growing sharper over time. After leading hardware and autonomy efforts at Lyft, Jonny returned to the space sector with a strong conviction: EO systems need to be built around the mission, not just the sensing capabilities. In many EO programs, engineers build the satellite first and figure out what it’s really good for later, resulting in a culture of chasing throughput, novelty, and resolution arms races. But in truly mission-critical applications, whether for climate tracking, wildfire intelligence, or national defense, that approach falls short. Stakeholders need the right measurements, at the right cadence, delivered through the right infrastructure. That means designing the mission from first principles, with the end-user at the center. It also means closing the loop between sensing, analysis, and action—designing not just spacecraft, but end-to-end systems. Muon Space was built to embody that philosophy from day one, and that mindset is shared across a founding team with deep, cross-domain experience.
Pascal Stang, Muon’s CTO, previously led engineering for Google X’s Project Loon, developing high-altitude platforms for global internet delivery. He later directed autonomy efforts at Lyft’s self-driving division. Paul Day, Muon’s VP of Spacecraft Production, built the original Skybox spacecraft and went on to lead satellite production at Terra Bella and Planet. He also played a key role in developingApple’s Emergency SOS via Satellite feature. Reuben Rohrschneider, Chief Mission Architect, was formerly Chief Engineer for Ball Aerospace’s MethaneSAT mission and brings deep expertise in satellite instrumentation and climate sensing. And anchoring the company’s scientific vision is Dr. Dan McCleese, Muon’s Chief Scientist and the former Chief Scientist at NASA’s Jet Propulsion Laboratory. A veteran of multiple planetary missions, including the Mars Climate Sounder, McCleese brings both scientific rigor and policy relevance to Muon’s climate and environmental programs. Guiding the company’s growth is Greg Smirin, Muon’s President, a veteran startup operator who has helped scale multiple companies, including several that went public..
Today, Muon has grown to a team of more than 100 drawn from top aerospace, autonomy, climate science, and national security organizations. That depth has enabled them to evolve into a full-capability organization as a mission design leader, a systems engineering powerhouse, and having deep fluency across both orbital and ground infrastructure.
A Full-Stack Platform for the Earth Intelligence Era
Delivering real-time, actionable Earth intelligence demands a rethink of how satellite infrastructure is architected, from isolated components into tightly integrated systems that move at the speed of software. Where traditional providers focus on building spacecraft or selling data, Muon has developed a vertically integrated architecture that spans simulation, spacecraft, middleware, sensing, control, and use-case specific analytics. This end-to-end model, what Muon calls Space-as-a-Service, is powered by a modular technology stack designed for rapid deployment, seamless integration, and continuous mission optimization.
MuSim: Digital Twin for Mission Design
Every mission begins in MuSim, Muon’s proprietary digital twin environment. It models orbital dynamics, spacecraft behavior, instrument performance, cloud connectivity, and latency across a full mission timeline. This allows teams to simulate thousands of design variations before hardware is ever built, surfacing tradeoffs, de-risking edge cases, and co-iterating with users to shape the right system from the outset. This simulation-first approach is what enables Muon to move fast without cutting corners. Three spacecraft generations have already been fielded in just three years, with every revision built atop the thousands of system-level design hours accumulated in MuSim.
MuSat + MuCore: Modular Spacecraft and Software-Defined Sensing
At the hardware layer, MuSat is Muon’s configurable spacecraft platform. It supports missions from 100-kilogram climate monitors to kilowatt-class systems tailored for national security. Its modular design allows core components to be reused and reconfigured across mission classes, preserving flexibility without compromising build timelines or manufacturing consistency. Powering these platforms is MuCore, a software-defined instrument controller capable of high-throughput electromagnetic data capture across modalities, including optical, IR, RF, hyperspectral, and more. MuCore is built not just to collect data, but to process and route it with real-time adaptability. Together, MuSat and MuCore form the physical and digital foundation of Muon’s in-orbit stack, scalable, reconfigurable, and built for mission diversity.
MuOS: Middleware for Seamless Space-to-Cloud Operations
Sitting between spacecraft and users is MuOS, Muon’s cloud-native middleware. It functions as a software-defined network layer, managing inter-satellite coordination, ground station access, data routing, and tasking. MuOS eliminates many of the manual integration bottlenecks that slow legacy missions, replacing them with software primitives that scale automatically with mission complexity. This layer transforms spacecraft from passive sensors into intelligent edge nodes, capable of routing, prioritizing, and communicating in real time across a global operational mesh.
MuDash: Command and Control for the Platform Era
On the user-facing side, MuDash provides a modern, browser-accessible interface for mission operations. Through MuDash, operators can schedule tasking, monitor system health, issue software updates, and access telemetry in real time. More than a dashboard, MuDash abstracts away the complexity of spacecraft control and provides users with direct access to system behavior, whether through GUI or API. This is critical to Muon’s space-as-a-service promise. It puts space infrastructure in the hands of those who need it, without requiring them to become day-to-day satellite operators.
What makes this stack of technology powerful isn’t just modularity—it’s the feedback loop. Data from MuDash flows into MuSim to refine future system designs. MuCore telemetry informs in-orbit performance tuning. MuOS enables constellations to act not as independent assets, but as a coordinated, software-defined sensor web. Because Muon owns the full stack, iteration can occur across the entire lifecycle. Hardware evolves with software. Missions are treated not as one-off programs, but as part of a continuously improving product pipeline. Muon understands that accelerating impact isn’t just about building and launching satellites quickly, it’s about getting the foundation right and this is what enables real-time, operational understanding of our planet at the scale the world now requires. It’s a blueprint not just for EO, but for a new generation of intelligent, adaptable orbital systems built for a resilient future.
Executing at the Speed of Orbit, for a Smarter Planet and Beyond
The journey to FireSat’s operational status illustrates Muon’s ability to move with urgency and precision. Following conceptual formulation studies led by Google and EDF in 2021-2022, Muon secured a $42M Phase I contract with the Earth Fire Alliance in 2024 to deploy the dedicated constellation. The first satellite, FireSat-0, launched in early 2025 and has since entered healthy on-orbit operations. It is currently undergoing commissioning, with initial data expected soon and deliveries to the National Reconnaissance Office (NRO). A major Phase II expansion is already in planning, with a substantial increase in satellite count and scope. While FireSat may be Muon’s most visible program to date, it is far from the company’s only proof point. In just a few years, Muon has quietly built a track record of delivering high-performance full-stack satellite platforms for an increasingly diverse set of customers across national security, climate monitoring, and next-generation communications.
One of the most prominent examples is the partnership with Sierra Nevada Corporation (SNC), where Muon was selected to build the first three satellites for SNC’s Vindlér RF sensing constellation. Announced in mid-2024, the program is expected to expand to more than twenty spacecraft, and reflects growing trust in Muon’s ability to deliver end-to-end space systems that meet specialized operational needs. That same approach is gaining traction across defense. In May 2025, Muon was awarded a Stage II contract with the NRO to validate the on-orbit performance of its electro-optical and infrared systems. The mission emphasizes persistent coverage and rapid revisit, capabilities enabled by Muon’s taskable, vertically integrated architecture. Additional defense awards include a $2.9M SBIR Phase II contract from the U.S. Space Force to advance multispectral cloud and weather imaging, as well as a 2023 award from the Defense Innovation Unit (DIU) for real-time space weather monitoring.
On the commercial side, Muon’s technology is increasingly sought after for novel sensing use cases. In 2023, the company launched a joint mission with Hydrosat to integrate a thermal infrared sensor for agricultural and climate analytics. A dozen more projects have extended Muon’s stack to support hyperspectral sensing, SAR imaging, and other advanced modalities, each reinforcing the configurability and breadth of its full-stack approach. These missions are anchored by the company’s Halo platform. In late 2024, Muon announced that they secured a $57M Series B and surpassed $100M in Halo-class contracts.
As more organizations turn to space for real-time awareness, whether for climate, security, or global connectivity, it becomes increasingly clear that speed, precision, and end-to-end ownership are non-negotiable. In this new landscape, Muon has become the partner of choice for missions that cannot afford to fail. Whether starting from a clean sheet or bringing an advanced payload, customers can focus on their mission while Muon delivers the rest. And while EO intelligence remains the foundation, the architecture is built to go further. The same vertically integrated stack can power space-based compute, resilient communications, precision timing, and direct-to-device IoT.
Sensors may change, form factors may shift, but the core principle remains: design for maximum mission alignment, build with user intent, and deliver reliable systems that perform exactly as expected. That’s the kind of future defined by trust and performance that Muon is propelling us all toward, and at Space Capital we couldn’t be prouder to have supported this extraordinary team from day zero.
