Seattle-based Constellation Space is rolling out its flagship software platform, which uses machine learning to streamline operations for fleets of satellites and the ground networks that serve them.

“We build models in-house for things like weather, attenuation, signal jamming, demand … and synthesize all of that to help operators understand what their ground contact plan looks like in the future, and how to optimize their fleets,” CEO Kamran Majid told GeekWire.

The Constellation Platform is tailor-made for a growing number of satellite constellations in low Earth orbit. Keeping track of thousands of satellites — and the best opportunities for downlinking data from each of those satellites — is becoming an increasingly complicated task.

Constellation Space, which was founded last year, draws upon Majid’s experience as a NASA engineering intern and as a software engineer working on SpaceX’s Starlink satellite network.

“At Starlink, I saw firsthand how manual mission-control processes break down as constellations scale,” he said. “I started Constellation to help smaller teams get ahead of link failures instead of spending their time rebuilding pass plans and recovering from outages.”

Majid’s fellow co-founders include chief technology officer Raaid Kabir, a veteran of Jeff Bezos’ Blue Origin space venture, and Omeed Tehrani and Laith Altarabishi, who were previously software engineers at Capital One. “We’re a team of six people right now,” Majid said.

One year into its existence, Constellation Space has already gained traction. 5 million in a seed round backed by Y Combinator, Nvidia, OpenAI, Samsung Next and other investors.

Source link