Sift Raises $42M to Build the Intelligence Layer for Rockets, Satellites, and Defense Systems

Every rocket launch generates a blizzard of data. A Falcon 9 streams up to 100,000 sensor parameters per second during flight. A satellite constellation produces petabytes of telemetry across thousands of spacecraft. A hypersonic defense system generates millions of data points from hundreds of sensors in milliseconds. And until recently, the tools used to make sense of all that data were either built internally by companies like SpaceX — or they simply did not exist.

Sift, a company founded by two former SpaceX engineers who built those internal tools, has raised a $42 million Series B led by StepStone to commercialize the data infrastructure layer for mission-critical hardware. GV (Google Ventures), Riot Ventures, Fika Ventures, and CIV also participated, bringing Sift's total funding to $67 million at a post-money valuation of $274 million.

The Problem: AI Cannot Control What It Cannot Understand

The thesis behind Sift is deceptively simple. As artificial intelligence shifts from processing text and images to controlling physical machines — rockets, satellites, autonomous vehicles, defense platforms — a fundamental infrastructure gap has emerged. These machines are essentially computers wrapped in steel, generating millions of data points per second from hundreds of sensors. But AI cannot interpret any of that data without structure.

Manual monitoring works when engineering teams can dedicate thousands of hours to a single vehicle. SpaceX can afford to build custom telemetry tools for Falcon 9 and Starship because the per-unit economics justify the investment. But most aerospace and defense companies do not have SpaceX's software engineering resources. And as the industry moves from operating individual vehicles to managing fleets of hundreds or thousands of machines, manual approaches break down entirely.

What Sift Actually Does

Sift provides a unified data platform with three pillars: storage, real-time analytics, and visualization. The platform ingests high-frequency sensor data from hardware systems — telemetry from rocket engines during test firings, performance data from satellites in orbit, sensor readings from autonomous vehicles during operation — and organizes it into a structured, queryable format that both human engineers and AI systems can use.

Some of the vehicles Sift works with have more than 1.5 million sensors streaming data concurrently, across multiple formats and timescales. The platform handles ingestion, storage, time-series alignment, anomaly detection, and fleet-wide comparison — the unglamorous but essential plumbing that makes AI-driven hardware operations possible.

The Customer Base

Sift's customer roster reads like a who's who of next-generation aerospace. United Launch Alliance (ULA) uses Sift for telemetry across its launch vehicle programs. Astranis, the geostationary broadband satellite company, relies on Sift for spacecraft operations data. K2 Space, building large utility satellites, and Parallel Systems, developing autonomous rail vehicles, are also customers. Additional undisclosed enterprise defense programs round out the portfolio.

The diversity of the customer base is strategic. By serving launch vehicles (ULA), satellites (Astranis, K2 Space), ground transport (Parallel Systems), and defense systems, Sift is positioning itself as a horizontal platform rather than a space-specific tool. This cross-sector approach mirrors the playbook of successful enterprise infrastructure companies like Datadog and Splunk, which built horizontal observability platforms for software before expanding into specialized verticals.

The SpaceX Origin Story

Sift's founding story is inseparable from SpaceX. CEO Karthik Gollapudi spent four years at SpaceX leading the flight software operations team for the Dragon spacecraft. CTO Austin Spiegel worked at SpaceX for nearly five years, building manufacturing systems, test automation, and the telemetry system for Starlink. The two started thinking about the telemetry problem as far back as 2019, when they witnessed the failure of Boeing's Starliner OFT-1 demo mission — a failure that was ultimately traced to software and telemetry issues.

At SpaceX, they experienced firsthand both the power and the limitations of purpose-built telemetry tools. SpaceX's internal systems are extraordinarily capable but were built specifically for SpaceX's hardware, by SpaceX engineers, using SpaceX infrastructure. The rest of the aerospace industry had nothing comparable. Sift was founded in 2022 to close that gap — to give every hardware company the telemetry infrastructure that previously only SpaceX could build.

The Growth Plan

The $42 million will fund an aggressive expansion. Sift plans to nearly double its team from 70 to approximately 140 employees, with a focus on engineering talent. The company is moving from its original El Segundo office to larger headquarters in Marina Del Rey to accommodate the growth. The platform roadmap includes expanded support for AI-driven anomaly detection, fleet-wide health monitoring, and automated test analysis — capabilities that become critical as customers scale from single vehicles to constellations and fleets.

The valuation trajectory tells a story about market timing. Sift went from a $7.5 million seed round in November 2023 — raised largely from SpaceX alumni — to a $274 million valuation in roughly two years. That 36x valuation increase reflects both the company's execution and the market's growing recognition that hardware observability is becoming a must-have infrastructure category, not a nice-to-have analytics tool.

Why This Matters for the Space Economy

The space industry is undergoing a transition from artisanal to industrial. The first generation of commercial space companies — SpaceX, Blue Origin, Rocket Lab — built everything in-house, including their software tools. The next generation cannot afford to do the same. Companies building satellite constellations, lunar landers, orbital servicing vehicles, and space stations need enterprise-grade infrastructure tools that work out of the box.

Sift is betting that telemetry and hardware observability will become as essential to aerospace as cloud monitoring became to software companies. Every satellite operator, launch provider, and defense contractor generates massive volumes of sensor data. The ones that can structure, analyze, and act on that data — in real time, at fleet scale — will have a decisive operational advantage. Sift is building the platform to make that possible.

Frequently Asked Questions