Satlyt raises $8 million for AI software for satellites — TechCrunch

Satlyt, a startup headquartered in Sunnyvale, US, and Nairobi, has raised $8 million in seed funding to develop software that enables artificial intelligence models to run on satellites. The round was led by Houston-based venture firm Non Sibi Ventures, TechCrunch reports.

The company was co-founded by Rama Afullo, an engineer who previously worked in Google's cloud business and was briefly involved with SpaceX's Starlink satellite communications network in 2024. Satlyt does not plan to build its own spacecraft; instead, it is developing software designed to run on satellites made by different companies.

Data processing aboard the satellite

Satlyts's initial focus is improving spacecraft efficiency. Satellites typically depend on ground operators to resolve problems, while transmitting data to Earth can be expensive and slow. The company proposes using AI directly onboard, including to respond to anomalies and process sensor readings.

More current news is available on the UA.News Telegram channel Telegram.

Earlier this year, Satlyt deployed Google DeepMind's Gemma model on a spacecraft operated by Momentus. According to TechCrunch, the model reduced the volume of transmissions related to onboard software error reports by more than 60%. Afullo said such efficiency could save hundreds of thousands of dollars per year for a single satellite.

The next mission and an orbital cloud

This week, Satlyt's software is set to launch on a SpaceX rocket alongside the first prototype of Google's Project Suncatcher. It will operate on a spacecraft from Indian startup TakeMe2Space, which develops computing equipment for satellites.

The mission involves NASA, which is paying to test cloud-computing protocols in space; startup Stellerian, which will test image-processing tasks for space observation; and TakeMe2Space. Next year, Satlyt plans to attempt to create a shared computing system linking two different satellites.