Sift, the El Segundo-based startup building a data infrastructure layer for autonomous machines, has closed a $42 million Series B round led by StepStone, with participation from GV (Google Ventures), Riot Ventures, Fika Ventures, and CIV. The company will use the capital to expand its engineering team and double down on the data-trust infrastructure that increasingly defines whether AI-controlled hardware — across space, defense, manufacturing, and autonomy — can be trusted to make safe decisions in the field.
The round positions Sift as one of the most well-capitalized infrastructure-layer companies in the rapidly consolidating defense-tech and commercial-space tooling space, and it underscores a broader investor thesis: that the next generation of autonomous systems will be defined less by the hardware itself and more by the data plumbing that keeps it reliable.
[HIGHLIGHT]Founded in 2022 by two SpaceX alumni who built monitoring systems for rockets and spacecraft, Sift sits at the intersection of two of the most powerful trends in industrial technology: the proliferation of mass-manufactured autonomous hardware and the operational reality that machine-learning systems are only as good as the data feeding them.[/HIGHLIGHT]
The Round in Detail
The $42 million Series B was led by StepStone, a global private markets investor with a growing presence in defense and dual-use technology. Existing and new investors GV (Google Ventures), Riot Ventures, Fika Ventures, and CIV all participated. The company did not disclose its post-money valuation.
[STATS]
Series B Raised | $42M
Lead Investor | StepStone
Co-Investors | GV, Riot Ventures, Fika Ventures, CIV
Headquarters | El Segundo, CA
Founded | 2022
Team Target | ~140 people
[/STATS]
CEO Karthik Gollapudi told Tectonic Defense that the team has actually already doubled since the round closed "a few months ago" — a signal that the company is in the middle of an aggressive scaling sprint rather than waiting for the announcement to start hiring. As part of that scaling, Sift is opening a San Francisco office to complement its El Segundo headquarters and expects the combined team to reach approximately 140 people.
What Sift Actually Builds
Sift's core product is a data infrastructure platform that sits between the raw telemetry coming off autonomous hardware and the engineering and AI systems that need to act on it. In the company's own framing, the software "automatically transforms raw sensor data into structured, queryable data that both engineers and AI systems can work with."
That description undersells what is actually a deceptively hard problem. As fleets of drones, satellites, USVs, rockets, and other autonomous systems scale into the thousands, the volume and variety of telemetry that operators must ingest, validate, and act on grows nonlinearly. Conventional time-series databases, observability tools, and homegrown data pipelines start to creak — and any data-quality issue at the input layer cascades into every downstream model and decision.
Sift positions itself as the trust layer for that pipeline. The platform validates whether the data feeding AI systems is reliable and flags when it isn't, so that downstream autonomy doesn't make decisions on bad inputs.
[QUOTE]"What we want to double down on is our data infrastructure. When we came to this problem, we really saw it as a big data problem, because we were working at the company that built the largest satellite constellation in the world, built the largest rocket in the world. Obviously, in order to build those things and operate them, data is constrained, essentially." — Austin Spiegel, CTO and Co-Founder, Sift[/QUOTE]
The Founders
Sift was founded in 2022 by Karthik Gollapudi (CEO) and Austin Spiegel (CTO), both SpaceX alumni who spent years building monitoring systems for rockets and spacecraft. That background matters: SpaceX's manufacturing and operations cadence — building reusable rockets at a rate that previously seemed impossible, and operating the world's largest satellite constellation — produces a class of telemetry and operational data problems that very few engineering organizations have encountered firsthand.
Both founders saw, from the inside, how brittle the data pipelines underneath autonomous fleets can be at scale. Sift was founded specifically to abstract that infrastructure away — so that operators of autonomous hardware can spend their engineering effort on the things that differentiate them, rather than reinventing the trust layer beneath their telemetry.
Customer Base: From Launch Vehicles to Drones
Sift has assembled a customer roster that reads like a snapshot of the most active companies in commercial space and defense autonomy:
- **United Launch Alliance (ULA)** — One of the most established launch providers in the U.S.
- **Astranis** — Operator of dedicated small GEO communications satellites
- **K2 Space** — Builder of large, high-power satellite buses for next-generation missions
- **Parallel Systems** — Autonomous freight rail platform applying autonomy to ground transport
- **CX2** — Defense autonomy company
- "Companies that build drones" and undisclosed enterprise defense programs
- Maritime customers operating in the unmanned-surface-vessel category
Gollapudi told Tectonic Defense that customers continue to choose Sift because of "best in class data infrastructure for this problem with real-time data ingestion, querying and cost effectiveness." That breadth of customer types — launch, satellite, rail, drone, maritime, defense — also explains why investors are willing to underwrite the company at this stage. Sift is not narrowly a space company or narrowly a defense company. It is an infrastructure layer for autonomous hardware, and the addressable market expands with every new category of physical system that becomes software-defined.
[QUOTE]"As people iterate faster, they can build things more reliably. They get things to market faster. And that's very much what we see. Because if you have limited capital and you have something that there's already so many risks, but building your hardware, we can take on that software risk off the table." — Karthik Gollapudi, CEO and Co-Founder, Sift[/QUOTE]
Why It Matters Now
The timing of Sift's Series B reflects three converging forces in the broader autonomous-hardware market:
1. **AI-controlled hardware is multiplying.** Drone swarms, satellite mega-constellations, autonomous ground systems, and maritime unmanned platforms are no longer rare experiments — they are production deployments. Each one generates more telemetry than the last, and each is increasingly making operational decisions in software.
2. **Bad data is the bottleneck.** Most of the high-profile failures in autonomy — across both commercial and defense applications — trace back to data quality, sensor fusion, or telemetry-pipeline issues rather than to the underlying ML models. Trustworthy autonomy requires trustworthy data.
3. **Defense is buying infrastructure, not just platforms.** The Department of Defense and intelligence community are increasingly investing in software infrastructure that improves the operational reliability of autonomous systems they buy from many vendors, rather than betting on a single end-to-end stack.
[HIGHLIGHT]Investor capital is increasingly flowing to the unglamorous-but-critical infrastructure beneath autonomy — data ingestion, validation, observability, and trust — rather than to the autonomy itself. Sift is one of the clearest expressions of that thesis at the Series B stage.[/HIGHLIGHT]
What the Money Funds
The $42 million is targeted at engineering scale. In the company's own words, the capital will go toward expanding the engineering team to "build the infrastructure layer that AI-controlled hardware systems will depend on across space, defense, manufacturing, and autonomy."
Specifically, Sift plans to:
- Open a San Francisco office to complement the El Segundo headquarters
- Roughly double the team to approximately 140 people (much of that already underway since the round closed)
- Continue investing in real-time data ingestion, querying performance, and cost efficiency
- Expand integrations across additional autonomous-hardware verticals
The Bottom Line
Sift's $42 million Series B is a meaningful data point in two overlapping stories. The first is the maturation of the commercial-space and defense-autonomy markets, where infrastructure-layer companies are now attracting Series-B-scale capital on the strength of real customer revenue rather than thesis-driven bets. The second is the broader ascendance of "trust-layer" tooling — companies whose entire job is to make sure that AI-controlled hardware doesn't act on bad data.
If Sift can execute on the hiring, the product expansion, and the customer growth implied by this round, it has a credible path to becoming default infrastructure for any organization operating fleets of autonomous machines — across space, defense, and far beyond.
[FAQ]
Q: How much did Sift raise and who led the round?
A: Sift raised a $42 million Series B led by StepStone, with participation from GV (Google Ventures), Riot Ventures, Fika Ventures, and CIV. The post-money valuation was not disclosed.
Q: Who founded Sift?
A: Sift was founded in 2022 by CEO Karthik Gollapudi and CTO Austin Spiegel, both SpaceX alumni who built monitoring systems for rockets and spacecraft.
Q: What does Sift do?
A: Sift builds a data infrastructure layer for autonomous machines. Its software automatically transforms raw sensor data into structured, queryable data that both engineers and AI systems can use, while validating whether the data is reliable and flagging issues — effectively a "trust layer" for autonomy.
Q: Who are Sift's customers?
A: Publicly named customers include United Launch Alliance, Astranis, K2 Space, Parallel Systems, and CX2, alongside additional drone manufacturers, maritime operators, and undisclosed enterprise defense programs.
Q: Where is Sift based and how big is the team?
A: Sift is headquartered in El Segundo, California, and is opening a San Francisco office. The company plans to roughly double the team to approximately 140 people, with much of that growth already underway since the round closed.
Q: How is Sift different from general observability tools?
A: Generic observability and time-series tools are designed for software systems. Sift is purpose-built for the telemetry, sensor data, and operational decision-making patterns of autonomous physical hardware — where data integrity directly affects whether a drone, satellite, or vehicle behaves correctly.
[/FAQ]