Satlyt has raised $8 million in seed funding to expand software that could help satellites process more data in orbit and reduce the cost and delays linked to sending large amounts of information back to Earth.
The Sunnyvale- and Nairobi-based startup develops software that allows artificial intelligence models and other computing workloads to run directly on satellites. The technology could help satellite operators make better use of computing resources already in orbit while creating new opportunities for companies to use satellites as managed computing platforms.
The funding round was led by Houston-based Non Sibi Ventures. Satlyt was co-founded by Rama Afullo, a former product manager at Google and SpaceX. He says the company’s longer-term goal is to allow multiple satellites to work together when handling large computing tasks.
Rather than building its own spacecraft, Satlyt is developing software that can operate across satellites owned by different companies. Afullo compares the model to VMware and Snowflake, which provide technology that operates across existing infrastructure.
“The folks like SpaceX who are doing orbital data centers, if they are the iPhone, we’ll build Android as a horizontally integrated, open ecosystem,” he said.
Software Designed to Reduce the Cost of Moving Data
Satlyt has already tested its software on two demonstration missions. The technology is designed to allow satellites to process information before it is transmitted back to Earth. This can reduce both the amount of data that needs to be sent and the time required to move it.
Earlier this year, Satlyt deployed Google DeepMind’s Gemma AI model onboard a Momentus spacecraft. The test reduced a transmission involving software errors by more than 60%. According to Afullo, the savings could reach hundreds of thousands of dollars per satellite each year.
The company is now taking the technology into another live mission. Its software is launching this week on a SpaceX rocket onboard a spacecraft built by Indian startup TakeMe2Space, which develops computing hardware for satellites.
The mission has three customers. NASA is paying to test protocols for cloud computing in space. Stellerian, a space-surveillance startup that tracks objects in orbit, will test image-processing workloads. TakeMe2Space will also use the mission to demonstrate that its satellite can host software from other companies.
The launch comes as major technology companies increase their interest in orbital computing. SpaceX is developing its own approach to data centres in space, while Google is advancing Project Suncatcher, its space data centre initiative.
Satlyt, however, is taking a different approach. The company is not relying on large orbital data centres being built before its technology can be useful.
“We don’t need data centers in space for Rama to be wildly successful, right?” Non Sibi partner Kent Lucas told TechCrunch. “It can just be driven by the number of satellites going up.”
Funding to Build a Shared Computing Network
Satlyt plans to use the new funding to develop a shared computing system that can operate across two satellites. The company plans to attempt the project next year.
If successful, the system would move Satlyt closer to operating a cloud computing network in orbit as a third-party provider. It would also demonstrate how computing resources from separate spacecraft can be combined to handle larger workloads.
Afullo said the company ultimately wants to turn satellites into platforms that can generate additional revenue for their owners.
“The way to think about it is that we turn your satellite into a revenue-generating managed service,” he said.
The $8 million round also reflects investor interest in software and infrastructure supporting the emerging space-computing market. The funding ranks among the top 5% of US enterprise software seed deals by historical size, according to the information provided.
With relatively few high-powered GPUs currently operating in space, Satlyt is focusing on software efficiency and the growing number of satellites being launched. Its approach could allow existing spacecraft to support more computing workloads without requiring operators to build entirely new orbital infrastructure.
