A proliferating LEO constellation designed to dramatically increase global observational density and maximize the potential of AI-driven predictions
Tomorrow.io, the world’s leading resilience platform, announced DeepSky, the world’s first AI-native space-based atmospheric and ocean sensing network. The network is designed to provide real-time, continuous observations of Earth’s atmosphere and oceans, enabling faster and smarter global decision-making and response. This is Tomorrow.io’s second constellation and was announced just one week after the first constellation was fully deployed with Ka-band radar and microwave sounder.
As global weather companies look to the next decade, DeepSky positions Tomorrow.io at the forefront of a new generation of weather infrastructure designed to be AI-native, operationally resilient, and commercially scalable.
Artificial intelligence is transforming weather forecasting, making it much more accurate, faster, and more efficient. However, as models have advanced, predictive performance has become increasingly constrained not by algorithms or computational power, but by the global observing system itself. AI systems rely on high-density, high-frequency, and diverse observations that today’s satellite infrastructure cannot consistently provide.
Also read: AiThority interview with Zohaib Ahmed, co-founder and CEO of Resemble AI
DeepSky is built to fill this gap. It represents a new class of commercial weather infrastructure designed to provide the temporal density and observational diversity required by modern forecasting systems.
DeepSky is a proliferating low-Earth orbit (pLEO) constellation of high-performance satellites, each carrying multiple high-impact unique instruments. Because this constellation operates at scale, it can achieve dramatically higher revisit rates than traditional systems, speed up global and regional model update cycles, and improve predictions of rapidly evolving extreme weather events. DeepSky is designed to complement, rather than replace, the government’s existing GEO and LEO assets, extending their value through higher cadence and new sensing methods.
Key design principles include:
- Multimodal sensing on all satellitesspanning most of the electromagnetic spectrum available for atmospheric and ocean observations.
- Global coverage with high return visitssignificantly increasing temporal density in parallel with existing GEO and LEO systems
- A new class of sensorspreviously limited to bespoke science missions due to cost and revisit constraints
DeepSky combines scale and capabilities to enable observation cycles that were previously unrealistic, while significantly reducing cost per scan and cost per impact.
DeepSky is designed to support all operational weather users, including commercial weather agencies, severe weather and hurricane forecast centers, defense and national security agencies, and international partners.
“Operational resiliency now depends on treating atmospheric data with the same rigor as other mission-critical infrastructure.” Nikhil Ahuja, Senior Director, Planning and Supply Chain, said: On Amazon. “Advances in sensing and faster refresh frequencies enabled by DeepSky will create a new class of AI-driven decision-making systems that are more adaptive and localized. This evolution will define the future of the world’s largest operations.”
The constellation is intended to meet and significantly exceed the fundamental observation requirements of major government agencies and international customers, while remaining adaptable to evolving mission priorities and emerging applications.
By providing frequent, diverse, and globally consistent observations, DeepSky enables:
- Faster update cycles for global and regional models
- Improving predictions of rapidly evolving extreme weather events
- New AI-native applications not possible with today’s observation densities
“Modern supply chains can no longer rely on static plans or historical averages. True resiliency comes from continuously sensing operating conditions and translating that information into network-wide decision-making.” BNSF CTO Matt Garland said: “DeepSky represents a meaningful step toward defining a new category of what is possible as agent AI becomes a critical part of unified operational context planning and creation.”
The constellation builds on Tomorrow.io’s proven experience designing, launching and operating commercial weather satellites and providing operational data to customers around the world. Development is underway with a clear path to adoption and expansion.
Also read: The Death of Surveys: Automating RFP Responses with GenAI
[To share your insights with us, please write to psen@itechseries.com]
