Satlyt raises $8M to bring AI to satellites

Who is behind Satlyt

Satlyt was founded by Rama Afullo, a Kenyan American engineer who previously worked in Google’s cloud computing business and had a brief stint at SpaceX’s Starlink communications network in 2024. Afullo left SpaceX to launch Satlyt, which is now headquartered in Sunnyvale, California, and Nairobi.

According to TechCrunch, Afullo’s background in cloud computing and satellite communications informs Satlyt’s approach to enabling AI workloads in orbit. The company’s software aims to abstract the complexities of satellite infrastructure, similar to how cloud platforms manage underlying hardware for terrestrial applications.

Funding and investor rationale

Satlyt has raised $8 million in a seed round led by Non Sibi Ventures, a Houston-based venture capital firm. The round marks Non Sibi Ventures’ first investment in orbital computing, a sector gaining traction as satellite operators seek to reduce costs and improve efficiency.

Bernard Harris, a partner at Non Sibi Ventures and a former NASA astronaut with over 20 years of experience, cited Satlyt’s software-focused model as a key reason for the investment. According to TechCrunch, Harris emphasized that Satlyt avoids the capital-intensive challenges of building space data centers, instead leveraging existing satellite infrastructure.

How Satlyt’s technology works

Satlyt’s software enables AI models to run directly on satellites, reducing the need to transmit raw data back to Earth for processing. This approach can cut downlink transmission sizes by over 60%, according to demonstrations cited by TechCrunch.

The company’s software has already been tested on two missions, including one where Google DeepMind’s Gemma AI model was deployed on a spacecraft operated by Momentus. This trial demonstrated the potential to save satellite operators hundreds of thousands of dollars annually by processing data in orbit.

Satlyt’s next launch will take place on a SpaceX rocket alongside Google’s Project Suncatcher prototype. The mission involves a spacecraft built by TakeMe2Space, an Indian startup, and is funded in part by NASA, which is testing protocols for cloud computing in space.

Satlyt’s long-term vision

Satlyt aims to create an open ecosystem for orbital computing, enabling multiple satellites to share computing workloads in space. Afullo has ambitious goals: the company hopes its software will be live on 20% of satellites by the end of the decade.

The startup also plans to develop a shared computing system spanning multiple satellites, which could further reduce costs and improve efficiency for operators. Competitors like Starcloud and Cowboy Space Company are exploring similar concepts, but Satlyt’s software-only approach sets it apart.

Afullo predicts that virtually every spacecraft builder will include GPUs in their satellites by the end of the decade, creating a growing market for Satlyt’s technology.

Early adoption and partnerships

Satlyt has already gained traction with early customers and partners. NASA is funding a mission with TakeMe2Space to test Satlyt’s software for cloud computing protocols in space. Another customer, Stellerian, a startup focused on space surveillance, plans to test image processing workloads on Satlyt’s platform.

The company’s partnerships extend to Momentus, which hosted Satlyt’s software on a previous mission, and TakeMe2Space, which is providing the spacecraft for the upcoming launch.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

Satlyt’s funding and technology highlight a shift in the space industry toward software-defined capabilities, particularly AI, as a way to unlock new efficiencies. For founders, this story underscores how startups can leverage existing infrastructure—whether satellites or cloud platforms—to build scalable solutions without bearing the full cost of hardware development.

The orbital computing model also reflects broader trends in edge computing, where data processing happens closer to the source, reducing latency and costs. Satlyt’s open ecosystem approach could lower the barrier for satellite operators to adopt AI, much like how Android democratized mobile development. However, the challenge will be convincing operators to trust third-party software for critical missions—a hurdle that will require robust security, reliability, and performance guarantees.

For investors, Satlyt’s focus on software rather than hardware may be appealing, as it avoids the capital-intensive risks of building physical infrastructure. The involvement of a former NASA astronaut as an investor also signals credibility in a sector where technical expertise is paramount. Still, the success of orbital computing hinges on broader adoption of AI-capable satellites, which remains an emerging market.

Key takeaways

• Founders should note Satlyt’s focus on solving a specific pain point: reducing costly downlink transmissions by enabling AI processing in orbit.
• The orbital computing model mirrors cloud computing architectures, suggesting opportunities for startups to abstract infrastructure complexities for space-based applications.
• Satlyt’s open ecosystem approach could lower barriers for satellite operators to adopt AI, similar to how Android democratized mobile development.
• Investors may see potential in startups that avoid capital-intensive hardware development, as Satlyt does by focusing solely on software.
• Early customer traction, even with niche players like Stellerian and NASA, can validate demand and attract venture capital in emerging sectors like space tech.

FAQ

Why is running AI on satellites valuable?

Running AI on satellites reduces the need to transmit large amounts of raw data back to Earth for processing. This can cut downlink transmission sizes by over 60%, saving satellite operators hundreds of thousands of dollars annually in bandwidth and operational costs.

What makes Satlyt’s approach different from competitors?

Satlyt focuses solely on software, enabling AI workloads on existing satellite infrastructure without requiring operators to build their own spacecraft or data centers. This mirrors the cloud computing model, where companies like VMware and Snowflake abstract infrastructure complexities for users.

Who are Satlyt’s customers and partners?

Satlyt’s customers include NASA, which is testing its software for cloud computing protocols, and Stellerian, a space surveillance startup. Partners include TakeMe2Space, an Indian startup providing spacecraft for an upcoming launch, and Momentus, which hosted Satlyt’s software on a previous mission.

What are the risks for Satlyt?

Satlyt’s success depends on broader adoption of AI-capable satellites, which is still an emerging market. The company must also convince operators to trust its software for critical missions, requiring robust security, reliability, and performance. Additionally, competition from companies like Starcloud and Cowboy Space Company could intensify.

Related on Lazyfounder

• Google
• SpaceX
• Starlink
• NASA
• More startup news