Caelus Space

**Product and the concrete problem it solves.** Caelus Space addresses a failure mode that is expensive precisely because it develops out of sight: roads, railways, bridges, dams, retaining walls, buildings, and pipelines can shift or lose ground before a visible crack or catastrophic collapse gives an operator a clear warning. Traditional monitoring is often local, periodic, and expensive to install, requiring survey crews, boreholes, ground sensors, or inspections at selected points. That leaves owners with a coverage problem across long corridors and distributed asset portfolios. Caelus presents a satellite-based alternative that monitors large areas repeatedly and turns movement into engineering and insurance risk signals. Its public product pages list building settlement, infrastructure subsidence, retaining-wall stability, sinkhole development, cliff-wall stability, vegetation and fire locating, water bodies, soil moisture, and surface models. The practical value proposition is earlier intervention: prioritize a site for inspection, maintenance, traffic restrictions, drainage work, or underwriting review while there is still time to prevent a larger loss.

**Core technology and how it actually works.** Caelus says its algorithms process multiple satellite-sensor inputs together with structural data, historical data, and regulations or industry standards. The resulting AI engine applies machine-learning and image-processing methods to identify movement patterns and convert them into digital reports, local site data, compliance alerts, and infrastructure-management outputs. Its infrastructure page claims monitoring at 2 mm precision every six days, while the company describes its broader model as combining subsurface data with satellite monitoring. Those are meaningful technical claims, but they should remain attributed to the company until independently benchmarked across different soils, slopes, vegetation, atmospheric conditions, asset types, and satellite passes. The architecture appears to be a software and analytics layer over remote-sensing data rather than a satellite operator building its own constellation. That reduces capital intensity and can support broad coverage, but it also makes data access, calibration, change detection, geospatial registration, and expert interpretation central to product quality. The engineering output matters as much as the model: an owner needs a defensible location, magnitude, trend, confidence level, and recommended next action, not merely an anomaly heat map.

**Market, customers, and go-to-market.** Caelus is aimed at organizations that manage geographically dispersed assets or carry financial exposure to their failure. The clearest target groups visible in public materials are road and rail authorities, bridge and dam operators, utilities and pipeline owners, construction and engineering firms, insurers, and public-safety organizations. The company describes a B2B model and emphasizes an ability to monitor large networks without placing a physical sensor on the ground. That supports several go-to-market paths: sell a recurring monitoring and reporting service to an infrastructure owner; provide risk evidence to an insurer; work through civil-engineering partners that already advise asset managers; or expose results through a platform or API inside existing asset-management workflows. Its published projects include city subsidence in Cour-Cheverny, a sinkhole and ground-undulation case on Israel’s Road 431, retaining-wall stability in Beit Shemesh, water-body analysis after Typhoon Rai in the Philippines, and infrastructure mapping in Tel Aviv. These examples show a geographically varied use-case portfolio, although the public record does not name paying customers, contract sizes, renewal rates, or the exact commercial status of each project.

**Traction, funding, and third-party validation.** Caelus has several public signals that are stronger than a concept page but still require disciplined interpretation. IVC lists Caelus.Space Ltd. as a Kfar Saba company with a seed round dated September 14, 2025; the amount and investor identity are not disclosed in the accessible record. Israel’s public company-information portal lists CAELUS.SPACE LTD as an active Israeli private company incorporated on April 10, 2025, which corroborates a very young corporate entity. The company’s LinkedIn profile states that it has been selected by the Israel Insurance Ecosystem as its 2026 Startup of the Year and says it already helps clients save hundreds of millions of dollars. It also reports monitoring more than 20,000 km of critical road infrastructure, with additional deployments underway. Those operating metrics and awards are company-reported or ecosystem-reported, not audited performance evidence. The 10D Israeli Physical AI map independently includes Caelus in its early-stage space category and describes it as using satellite imagery and machine learning to detect ground-movement hazards. No public source reviewed here establishes revenue, a priced institutional round, a patent portfolio, regulatory certification, or an independently validated accuracy benchmark.

**Founders and team background.** Caelus’s public trust page identifies Amit Hellman as CEO and co-founder, Dror Sharf as CTO, and Yoav Ben Or in insurance business development. Hellman is described as a third-generation building contractor with a PhD in geoinformatics and geotechnical engineering, more than a decade of professional experience, dozens of published academic papers, and teaching responsibilities in civil-engineering courses on infrastructure monitoring and safety. That combination is unusually well matched to the customer problem: it links construction failure modes and engineering workflows to geospatial analytics rather than treating satellite imagery as a generic data product. Sharf is described as holding a master’s degree in physics and computer science, with more than a decade of professional experience and a background solving complex technical challenges. Ben Or’s role suggests the team understands that insurance adoption depends on underwriting, claims, loss prevention, and evidentiary workflows rather than only on sensor performance. The public team footprint remains small, and the sources do not disclose prior exits, a full engineering roster, named academic collaborators, or patent ownership. Those gaps matter for assessing whether the company can move from a technically credible founder-led product to a repeatable enterprise deployment organization.

**Competitive dynamics.** Caelus operates in a layered market rather than an empty category. ASTERRA, formerly Utilis, uses satellite synthetic-aperture radar and AI for underground water-leak and infrastructure analysis; SkyGeo and SatSense offer InSAR-based monitoring for infrastructure and ground movement; ICEYE supplies high-frequency synthetic-aperture-radar imagery and analytics; and engineering or measurement incumbents such as Fugro and Hexagon can combine surveys, field instrumentation, geospatial software, and consulting relationships. The relevant substitute is not only another startup. Owners may continue with periodic inspections, GNSS instruments, total stations, piezometers, fiber sensing, or engineering firms that bundle ground monitoring into a larger project. Caelus’s potential edge is the combination of geotechnical expertise, multi-layer subsurface and satellite data, engineer-ready reporting, and broad coverage without installing a sensor at each location. That may improve the economics of screening a whole road or rail network before sending a crew to a small number of priority sites. The edge is conditional: competitors can access similar satellite data, customers may demand raw-data provenance, and a false negative on a critical asset can outweigh many successful low-risk alerts. Durable differentiation would require a proprietary labeled dataset, superior change-detection performance, calibrated uncertainty, workflow integration, and evidence that its alerts lead to measurable avoided loss.

**Defense, security, and resilience dual-use relevance.** Caelus is best classified as resilience-first dual-use technology. Its commercial use is direct: insurers can use early ground-movement evidence for underwriting and loss prevention, while transportation authorities, utilities, construction firms, and asset owners can use it to prioritize inspections and maintenance. The same monitoring layer can support public safety and security-sensitive infrastructure by screening roads, bridges, dams, pipelines, rail corridors, ports, energy facilities, and other assets whose disruption could affect mobility, emergency response, or national continuity. In a defense context, satellite-based movement detection could be relevant to access routes, airfields, depots, border infrastructure, hardened sites, and post-event damage assessment; it could also help identify landslide or subsidence risk along military logistics corridors. The evidence does not show defense contracts, classified work, military fielding, or a government procurement award, so those applications are strategic adjacency rather than realized defense traction. The core capability still credibly spans civilian and security or resilience settings because the underlying problem is asset movement and geotechnical change, not a consumer-only workflow. Data sovereignty, secure APIs, auditability, and controlled dissemination would determine whether allied government users can adopt it.

**Growth stage, trajectory, and key diligence risks.** Caelus is an early-stage startup: the Israeli company was incorporated in 2025, IVC records a seed event, the Startup Nation Central ecosystem listing places it at pre-funding with 1-10 employees, and LinkedIn presents an 11-50 employee range that likely reflects a broader or more recent estimate. Its trajectory could be attractive if a single monitoring service becomes a repeatable layer for entire infrastructure networks and if insurance loss-prevention budgets create a faster commercial path than public-works procurement. The near-term diligence questions are substantial. (1) Validate the claimed 2 mm precision and six-day cadence by asset, terrain, satellite source, and weather condition, including false negatives and false alerts. (2) Separate monitoring capability from business impact: the reported savings and 20,000-km coverage need customer-level evidence, not only aggregate company claims. (3) Establish data rights, data residency, satellite-provider dependence, API security, and continuity during provider outages or geopolitical restrictions. (4) Test whether engineering customers will pay for alerts without mandatory field verification, and whether insurers accept the outputs in underwriting or claims decisions. (5) Clarify customer concentration, recurring revenue, gross margin, implementation effort, and the balance between software and expert services. (6) Confirm the founders’ ability to scale product, geospatial engineering, sales, and customer support while preserving scientific rigor. (7) Map the company’s patent, academic-license, and algorithmic trade-secret posture. Caelus has a specific strategic problem, a technically coherent approach, and early ecosystem validation, but its investment signal should remain conditional on independently demonstrated accuracy and repeatable paid deployment.