Interview | Satellites For Hire: SkyServe's Flash Missions Redefine Satellite Use In Earth Observation

Bengaluru: Not long ago, satellites collected data at a measured pace and delivered it to a limited set of users. Today, the need has fundamentally changed. Decisions around disasters, climate events, and security increasingly depend on real-time insight, not delayed analysis. At the same time, modern satellites produce far more data than can be economically sent to the ground. Bandwidth constraints, latency, and limited access to ground stations have become critical bottlenecks. Recent advances in space-grade computing are breaking this impasse, making reliable, on-orbit data processing practical for the first time—something that was simply not feasible a decade ago.

For over 60 years, space missions have followed the same rigid playbook: build a satellite, raise tens of millions, wait years for launch—and once it’s in orbit, live with whatever you built. Even a 10-minute experiment demands a 10-year asset. This approach has created a growing crisis in Earth observation: most satellites still download everything they capture—clouds, noise, unusable imagery—and decide what matters only after it reaches Earth. As data volumes explode, this model is becoming economically and technically unsustainable.

In response, SkyServe is redefining the model. Instead of treating satellites as single-purpose hardware, SkyServe treats them like gig workers—able to accept short, software-defined “flash missions”, execute them, and move on.

“Satellites are evolving from passive data collectors into active decision-making systems. On-orbit processing isn’t meant to replace ground systems—it’s about enabling faster action when every minute counts. SkyServe is creating the software layer that makes this shift safe, scalable, and economically sustainable,” explained Vinay Simha, Co-founder and CEO of SkyServe.

In a conversation with ETV Bharat, Vinay Simha highlighted why Earth observation economics are breaking, how onboard AI dramatically cuts bandwidth and time to insight, and why “flash missions” could redefine humanity’s use of space.

On-Orbit Processing: From Images to Insights

SkyServe enables satellites to process data in orbit, identify what’s relevant, and transmit only insights—not terabytes of raw images. Satellites stop being passive cameras and start becoming decision-making systems. No satellite is built for a single experiment; a developer who wants to count ships near Mumbai doesn’t launch anything—they upload a small Python script, run it on a passing satellite for minutes, get the answer, and the code disappears.

Simha explained, “A developer uses SURGE to upload a model, choose a time window and data source, and schedule the task. The model runs securely on STORM, SkyServe’s onboard engine, sending only insights to the ground. Strong isolation ensures no code impacts the satellite or other missions.”

The Role and Limits of Flash Missions

“Flash missions are constrained primarily by power, thermal limits, and mission prioritisation. While onboard computing capability is advancing quickly, satellites still operate within tight energy budgets. Orbital dynamics further limit when and where observations can occur, and security and safety requirements restrict what software may be deployed. As a result, today’s flash missions are best suited to short, high-value tasks—such as detection, filtering, and alerting—rather than sustained, continuous processing,” said Simha.

STORM is SkyServe’s hardware-agnostic edge computing platform that runs AI models directly onboard satellites, transforming raw sensor data into near real-time, actionable insights. It enables in-orbit pre-processing, smart data filtering, and AI inference across multispectral, SAR, RF, and IoT payloads, cutting latency and reducing dependence on ground stations. (Image Credits: SkyServe)

A satellite owner can hand over onboard compute to third-party workloads. Discussing how SkyServe aligns incentives, Simha said, “This creates new revenue without new satellites. SkyServe runs workloads in isolated, mission-safe environments. Operators control when, where, and how compute is shared—and get paid per use. The core mission always has priority, so risk stays low while returns increase.”

When hundreds or thousands of developers compete for limited orbital compute, SkyServe manages resources fairly and profitably.

“The main constraint is power and scheduling, not compute,” Simha said. “SkyServe allocates resources using priority rules—based on mission criticality, timing, and contracts—like airline slots: limited, scheduled, and priced accordingly.”

Real-World Impact

In-orbit computing has moved beyond experiments into real deployments, including recent validation with NASA’s Jet Propulsion Laboratory. Operators face mounting pressure to cut downlink costs, reduce latency, and extract faster insights. The result: the next phase of the space industry will be software-led, not launch-led.

When asked about quantitative benefits, Simha shared, “Operators typically achieve a 5–10× reduction in data volume, up to 3× lower downlink costs, and response times cut from hours to minutes. In Mission Matterhorn, onboard analysis delivered results in under two minutes while significantly reducing bandwidth usage. This approach is well-suited for detection, classification, and change analysis, but it does not replace full-resolution scientific processing or use cases where raw data must be preserved for regulatory or research purposes.”

If most satellites stop downlinking raw data and transmit only insights, the impact on ground infrastructure and the business model is profound. Simha explained, “Ground systems shift from heavy data processing to decision support and integration. Data marketplaces evolve from selling raw imagery to delivering alerts, indicators, and outcomes—reducing infrastructure costs and expanding Earth observation to users who never needed images at all.”

Receiving insights instead of raw imagery raises questions about reliability, especially for regulated or high-stakes applications like defence, climate monitoring, and maritime enforcement. Simha said, “Trust is built on traceability and validation. Each insight is tied to the underlying model, data source, and confidence metrics. For regulated use cases, raw data can still be archived or selectively downlinked for audit and review. In short, insights accelerate decision-making while preserving the ability to verify results when required.”

Security and Safeguards

Regarding potential risks of hijacking or misuse, Simha explained, “Users never have direct control of the satellite. All access passes through authenticated, encrypted, and permission-controlled layers. Compute workloads run in sandboxes, commands are validated, and mission systems remain physically and logically isolated. Even in the event of a breach, core satellite control remains unreachable.”

He further said that with SkyServe, satellites are no longer just orbiting cameras—they’re on-demand decision engines. By combining onboard intelligence, security, and flexible access, the future of space is fast, scalable, and software-driven.