The Threat Has Changed Shape
LeoLabs launched Delta, an AI-powered space domain awareness platform, on April 8, 2026, to deliver real-time conjunction assessment for low Earth orbit objects — a direct response to maneuverable adversary payloads that have compressed satellite operators' decision windows. The platform replaces LeoGuard, the company's earlier SDA solution, and sits atop a vertically integrated stack: a proliferated network of phased-array radars, a commercial object catalog exceeding 26,000 tracked objects, and an analytics layer that fuses radar measurements with third-party data. Delta translates raw orbital data into actionable alerts delivered directly to operators, automatically consolidating observations into a living picture of activity in LEO and very low Earth orbit.
The system harnesses LeoLabs' global radar network, including a radar in New Zealand that detected the Chinese spaceplane releasing a new object in June 2026, now supplemented by the transportable Scout-S class, to maintain persistent custody of high-interest objects. Scout-S, housed in a standard 20-foot ISO container, extends observation windows from seconds to several minutes per orbital pass. Its direct radiating array and modular S-band design enable uncued search, discovery, and track-while-scan modes that feed Delta's analytics engine in near real time. The first operational Scout-S, deployed in the Indo-Pacific in June 2026, began observing satellites within hours of activation and has since demonstrated persistent track custody of Chinese ISR satellites, the Chinese spaceplane, and objects as low as 230 kilometers.
Delta's conjunction assessment capability delivers up to 400% more frequent updates than the legacy LeoGuard system. That cadence matters when a maneuver decision must be made before the next radar pass. The platform also powers two refreshed commercial services: LeoLabs Pulse for collision avoidance and LeoLabs Trace for launch and early operations support. Both draw on the same AI-driven analytics backbone, meaning a commercial operator and a defense customer share the same underlying truth about the orbital environment.
"We're giving U.S. and Allied operators the ability to not only see what's happening in orbit, but to understand what it means and act when it matters most," CEO Tony Frazier said at launch. Senior Vice President of International Sales Maher Khoury added that demand is accelerating globally as more nations field sovereign LEO capabilities critical to their national security. The platform's architecture — sensor-agnostic, cloud-native, built for interoperability — positions it to ingest data from allied sensor networks as those partnerships mature.
The launch coincides with a projected surge in LEO population: upward of 70,000 operational satellites by 2030, a growing fraction from adversarial nations. In that environment, the ability to maintain persistent custody of maneuverable payloads becomes the differentiator between awareness and surprise. Delta is LeoLabs' bid to make that custody routine.
Contract Momentum Follows the Product
LeoLabs' U.S. government business has accelerated with Delta in market. The company secured a $20.68 million Prototype Other Transaction Agreement from the U.S. Space Force in August 2026 to deliver a modified Scout system for a new defense mission set. Under the award, LeoLabs will demonstrate its containerized Scout radar, now integrated with Delta's real-time conjunction assessment and sensor-fusion layer, for missions that demand persistent custody of maneuverable threats in contested environments. The OTA follows a Tactical Funding Increase award that supported the first operational Scout-S deployment in the Indo-Pacific, where the system did so and has since tracked Chinese ISR satellites, including Yaogan military reconnaissance platforms between 800 and 1,000 km, and detected an object as low as 230 km.
Space Force acquisition officials have emphasized that the threat architecture has shifted: periodic object detection and orbit prediction are no longer sufficient; the requirement is persistent custody, characterization, and timely operational awareness. Delta's AI-driven analytics, built to ingest radar, optical, and passive-RF measurements into a single real-time catalog, directly addresses that gap. The platform's ability to extend them, and to maintain track custody on non-cooperative launches and reusable spaceplanes, maps to the Space Force's stated need for a distributed sensing layer that can be rapidly repositioned in contested environments.
Frazier framed the momentum as the payoff of a decade-long commercial investment: "We've spent the past decade building and operating high-performance radar systems while investing in transportable, proliferated sensing to meet emerging mission needs. Together, we're demonstrating how proven commercial technology can accelerate the delivery of new capabilities for U.S. defense missions." The Scout-S radar, in such a container and transportable by land, sea, or air, is now serving as a real-world testbed for Delta's analytics at the U.S. Indo-Pacific Command's Valiant Shield 2026 exercise — a live demonstration that the combined sensor-and-software stack can plug into existing defense architectures without the years-long integration cycles typical of exquisite government systems.
The contract trajectory suggests the Space Force is buying the integration, not just the sensor. The June 2025 SpaceWERX selection to further develop next-gen Scout capabilities, the TACFI-backed Indo-Pacific deployment, and the August 2026 OTA form a progression from technology demonstration to operational prototype — each step funded because the AI-enabled data product reduces the latency between detection and decision. LeoLabs' first-party board data shows the company adding roles in strategic sourcing, manufacturing, and senior software engineering in the Washington DC metro area and Denver, consistent with a team scaling to deliver on defense programs.
Whether Delta alone tipped the evaluation boards is not publicly documented. But the platform's launch, the $20.68 million OTA, and the Space Force's own statements on persistent custody, distributed sensing, and rapid commercial insertion align precisely with what Delta and the Scout family now deliver together.
Rivals Respond Across Orbital Regimes
The competitive field is stratifying along orbital regime lines. LeoLabs dominates commercial LEO radar. Slingshot Aerospace, based in Windsor, Colorado, secured a $69.2 million SBIR Phase 3 award from the Space Force on July 15, 2026, a 4½-year contract under the Operational Test and Training Infrastructure program. The award is explicitly built around TALOS, short for Thinking Agent for Logical Operations and Strategy. Slingshot describes TALOS as an AI-powered training and strategy agent that acts as an autonomous virtual opponent during exercises. Rather than following a predetermined script, TALOS simulates how another spacecraft could maneuver, respond to an operator's actions, or interfere with a mission. The company said the system "models realistic spacecraft behaviors, generates strategic response options, processes vast amounts of complex data, and supports mission rehearsal across continuously evolving space scenarios." The Pentagon's interest reflects a shift toward using AI not only to analyze large data volumes but to improve how military personnel train for fast-moving operational scenarios where satellites can maneuver, interfere with communications, or approach other spacecraft unpredictably.
Slingshot's move extends beyond software. The company sold optical sensors to the United Kingdom (its first hardware deal) and is in talks with other nations to create or expand sovereign space-tracking capabilities. That hardware push mirrors the sensor-software integration LeoLabs has pursued with its own radar network. In 2024, Slingshot reported a 17 percent year-over-year spike in the average number of close approaches in LEO per satellite, a data point that underscores the operational urgency driving both companies' AI investments.
Anduril Industries, a defense technology company valued at $61 billion as of May 2026, Wikipedia reports, has been awarded over $289 million in Department of Defense and Homeland Security contracts since March 2026, USAspending.gov's data shows. Hiring data from Zero G Talent's board shows Anduril added 220 roles in the past seven days, including a Senior Principal Chief Engineer for Space Special Programs and a VP of Production for Mission Systems — signals of a hardware-heavy ramp.
Slingshot's board footprint is smaller: two salaried roles, both senior technical architects in the $175,000–$250,000 band, and zero new postings in the past week. That suggests a leaner, software-first scaling model versus Anduril's hardware-heavy approach. Both approaches are valid responses to Delta. The Space Force's own contract pattern, awarding LeoLabs $20.68 million for a transportable Scout radar demonstration while simultaneously funding Slingshot's AI training environment, shows the service is hedging across sensor types and analytics paradigms. The message to industry: owning the sensor is no longer enough; owning the AI layer that turns sensor data into decision-grade intelligence is what wins the next contract cycle.
A Rush-Hour Freeway in Orbit
The space economy is expanding at a clip that makes low Earth orbit feel less like a frontier and more like a rush-hour freeway. The World Economic Forum projects the sector will reach $1.8 trillion by 2035, up from around $600 billion in 2023. Satellite assets already in orbit represent hundreds of billions of dollars of exposed value. Those numbers are not abstract — they are the balance sheet that every collision risk threatens.
Congestion is the driver. Research published in Nature documents a 127-fold increase in LEO satellite population over five years. SpaceX alone had launched more than 10,000 Starlink satellites as of October 2025, with independent trackers placing 8,500–8,700 currently on orbit. OneWeb, Kuiper, Telesat, and Europe's IRIS² constellation are adding thousands more. Each new spacecraft raises the conjunction probability for every other operator. The same study estimates 29 tons of satellites will re-enter the atmosphere daily once mega-constellations reach steady state, and the launch cadence required to sustain them could inject soot equivalent to 7 million diesel dump trucks.
That debris environment creates a dual market: tracking what is up there, and removing what should not be. On the removal side, the European on-orbit servicing segment alone is forecast to exceed $5 billion by 2030, compounding at 11.5 percent annually. Demonstration missions are moving from slides to launch manifests: Astroscale's ELSA-M targets a 2026 launch for magnetic capture of prepared defunct satellites; the U.S. Space Force's Tetra-5 and Tetra-6 refueling tests, with hardware from Astroscale, Northrop Grumman, and Orbit Fab, are slated for 2026 and 2027; ESA-backed ClearSpace-1 aims to deorbit the PROBA-1 satellite in 2029.
| Market Segment | Current Valuation | Projected Valuation | Target Year | CAGR / Notes |
|---|---|---|---|---|
| Global space economy (WEF) | ~$600B (2023) | $1.8T | 2035 | ~9% |
| European on-orbit servicing | — | >$5B | 2030 | 11.5% |
The economics of single-use satellites are breaking. Replacing failed or end-of-life spacecraft at constellation scale imposes recurring capital expenditure that operators and national budgets are eager to avoid. On-orbit servicing (refueling, repair, module swap) promises to convert that capex into a managed service. By 2030, analysts expect a shift to an industrial, service-based orbital economy where satellites are maintained rather than discarded. A robust ecosystem of tugs, repair platforms, logistics hubs, and interoperability standards would make that transition stick.
For AI-enhanced SDA, the implication is direct: the sensor network must scale faster than the object count. LeoLabs' radar catalog already tracks 26,000+ objects persistently. As mega-constellations densify, the gap between what radars see once and what they can keep custody of widens. AI-driven sensor fusion, automated conjunction assessment, and predictive maneuver planning are not optional features — they are the only way to keep the catalog current at the tempo the market now demands. The companies that close that loop win the contracts; the ones that do not watch the addressable market slip past them.
Inside the Loop: How Delta Drives the Sensors
LeoLabs' Delta platform replaces the company's earlier LeoGuard system, marking a shift from periodic conjunction screening to continuous, AI-orchestrated sensor management. Where LeoGuard relied on scheduled radar tasking and post-pass correlation, Delta harnesses LeoLabs' vertically integrated proliferated radar network and its world-leading commercial object catalog to continuously monitor mission-relevant objects, detect and characterize anomalous events, and analyze patterns of activity. It automatically translates and consolidates orbital data into actionable alerts and critical insights delivered directly to operators.
Sensor fusion sits atop that architecture. Delta fuses LeoLabs' own S-band radar returns with third-party optical, passive-RF, and space-based measurements into a single probabilistic state estimate for each object. The fusion engine maintains track custody on maneuvering payloads, rocket bodies, and debris fragments, critical for VLEO objects that dip below 300 km and reappear minutes later, then collapses hypotheses once radar re-acquisition locks covariance.
Scout-S, the transportable 3D search radar deployed in the Indo-Pacific, demonstrates the architecture at the edge. Its array and modular S-band electronics support uncued search, search-to-track, and track-while-scan modes simultaneously. The radar's onboard compute enables advanced detection modes that flag high-interest objects before raw data leaves the container. That edge processing cuts downlink volume and lets the central scheduler treat Scout-S as a smart sensor rather than a dumb collector.
The distinction from LeoGuard is operational, not cosmetic. LeoGuard screened a catalog; Delta drives the sensors that build it.
Working in frontier tech? Zero G Talent tracks the openings: see every open Anduril Industries role, browse frontier tech jobs, openings at Slingshot Aerospace and LeoLabs, and the people building the field.