Sift, an El Segundo startup focused on building an intelligence layer for mission-critical machines, closed a $42 million Series B financing round led by StepStone with participation from GV (Google Ventures), Riot Ventures, Fika Ventures and CIV, bringing total funding to $67 million. The investment will expand Sift’s engineering team to build the infrastructure layer that AI-controlled hardware systems will depend on across space, defense, manufacturing and autonomy.
“We started Sift because the infrastructure needed for AI-controlled hardware didn’t exist,” said Karthik Gollapudi, chief executive of Sift, in a statement. “Software observability matured over two decades, but hardware companies still rely on spreadsheets and tribal knowledge. Sift provides the intelligence layer that lets AI interact with hardware as fluently as it interacts with code.”
As artificial intelligence shifts from processing text to controlling physical machines, a fundamental infrastructure gap has emerged between what AI can do and what it can actually operate in the physical world. Modern rockets, satellites, defense systems and autonomous vehicles are computers wrapped in steel, generating millions of data points per second from hundreds of sensors, but AI cannot interpret any of it without structure. Sift automatically transforms raw sensor data into structured, queryable data that both engineers and AI systems can work with.
Sift co-founders Karthik Gollapudi and Austin Spiegel
Gollapudi and co-founder Austin Spiegel built monitoring systems for rockets and spacecraft at SpaceX. They recognized that while software companies had robust data infrastructure, the hardware industry lacked equivalent tools. Without an observability infrastructure built for hardware, that challenge compounds as the industry shifts from managing individual prototypes to operating fleets of hundreds or thousands of machines.
Sift serves as a single source of truth across audio, video, logs and high-frequency telemetry. Engineering teams use it to diagnose anomalies faster, validate designs against real-world performance and maintain fleet-wide visibility from a single platform.
The company’s platform addresses a fundamental scaling challenge. Manual monitoring works when engineering teams can dedicate thousands of hours to a single satellite or autonomous vehicle. But operating constellations of hundreds of machines generating petabytes of data requires automation: fleet-wide observability, automated anomaly detection and unified visibility across thousands of systems.
“We’re not building one satellite to operate for 15 years. We’re building hundreds and running constellations for decades,” said Neel Kujur, co-founder and chief technology officer at K2 Space, in a statement. “The amount of data will be enormous. Sift will be critical in making operations seamless, automatically flagging out-of-bounds telemetry and helping us close the design loop by using real-world data to improve.”
Trusted by ULA, Astranis, K2 Space, Parallel Systems and undisclosed enterprise defense programs, Sift is already the system of record for companies building the next generation of machines. With this funding, Sift plans to nearly double its team from 70 employees, move into a larger headquarters in Marina Del Rey after outgrowing its El Segundo office and expand its platform to support the rising number of organizations deploying AI-controlled hardware at scale.
Information for this article was sourced from Sift.
link