PerCiv

**Product and the concrete problem it solves.** PerCiv is aimed at a specific blind spot inside industrial water treatment: operators can see pumps, pressures, flows, conductivity, and other plant-level signals, but generally cannot see the condition of each membrane inside a pressure vessel without interrupting operations or waiting for a failure pattern to become obvious. Reverse-osmosis and related spiral-wound membrane systems are mission-critical in desalination, semiconductor manufacturing, power generation, oil and gas, food and beverage, and pharmaceutical production. When a membrane fouls, degrades, tears, or develops a brine leak, the plant may respond by increasing feed pressure, starting a clean-in-place cycle, replacing a complete vessel, or accepting reduced output. Those interventions are expensive and can be mistimed: more pressure may consume energy without restoring useful throughput, a cleaning cycle may not work uniformly, and a vessel-level average can hide the individual element that is deteriorating. PerCiv's current product proposition is an external, non-invasive monitoring layer that collects granular data from the membrane environment without putting wetted sensors into the process. The company says this provides an operator with membrane-level visibility, including which membranes need replacement, when a cleaning cycle is appropriate and whether it worked, when extra pressure is wasteful, and whether a brine leak is occurring and where it is located. The commercial problem is therefore not simply water quality; it is the operating cost, uptime, energy consumption, and maintenance uncertainty created by a process that has historically behaved like a black box.

**Core technology and how it actually works.** Public sources describe a combination of proprietary fiber-optic sensing, a patented monitoring method, physics-based analytics, and artificial intelligence. Startup Nation Finder identifies the underlying sensing approach as tiny fiber Bragg grating, or FBG, sensors that can detect strain, temperature, high-frequency vibration, and acoustic signatures simultaneously. In general technical terms, an FBG encodes a reflected optical wavelength in a short periodic structure inside a fiber; mechanical strain and temperature shift that reflected wavelength, allowing an interrogator to infer changes without running electrical sensing elements through the water stream. PerCiv's public material does not disclose the exact sensor placement, optical interrogation design, sampling rates, calibration data, model architecture, or patent number, so the record should not treat those implementation details as proven. The useful engineering hypothesis is nevertheless coherent: externally collected optical signals can be combined with process conditions and a physical model of a pressure vessel to distinguish normal operating variation from fouling, damage, vibration, or leakage. Physics-based analytics can constrain an AI system with known relationships among pressure, flow, temperature, and membrane behavior, potentially making anomaly outputs more interpretable than a plant-wide black-box score. The website says the method works without intrusion, wetted sensors, or disruption, and that it can be retrofitted to existing units. If validated, that retrofit characteristic matters because membrane arrays are already installed capital assets and customers are unlikely to replace an entire treatment train merely to obtain better diagnostics. The key technical diligence questions are sensitivity at membrane-element level, false-positive and false-negative rates, performance across water chemistries and membrane types, optical interrogator cost, calibration drift, and whether the method remains reliable under vibration, scaling, biofouling, and changing feed conditions.

**Market, customers, and go-to-market.** PerCiv's market is broader than municipal desalination. The company and ecosystem sources identify process industries in which filtration membranes affect production continuity and product quality: water and wastewater treatment, desalination, semiconductor fabs, power plants, oil and gas facilities, food and beverage lines, and pharmaceutical operations. The company positions the solution as a rapid retrofit for millions of existing units and says fast installation and quick return on investment are central to its offer. That suggests a B2B enterprise sale beginning with a plant, membrane train, or critical process line, followed by expansion across a fleet of pressure vessels or geographically distributed facilities. Buyers may include water-plant operators, engineering-procurement-construction firms, membrane service providers, industrial utilities, and process owners whose cost of unplanned downtime is materially higher than the monitoring subscription or hardware package. The Israel Innovation Authority record classifies the target customers as enterprises and describes the technology as a predictive-maintenance platform for machinery and structures. The public website lists the Kinneret Innovation Center in Tzemach, Israel, as the contact location, which places the startup in an applied water and industrial innovation environment. No named paying customer, commercial contract, revenue figure, installation count, or channel partnership is publicly confirmed in the reviewed sources. The likely go-to-market challenge is that a monitoring vendor must prove economic value against an incumbent maintenance routine, integrate with SCADA or plant historians, and earn the trust of operators who cannot risk an unvalidated intervention in a live desalination or high-purity process. A staged pilot that correlates sensor outputs with membrane autopsies, cleaning records, energy use, and replacement timing would be the most credible path to conversion.

**Traction, funding, and third-party validation.** PerCiv has more public validation than a purely anonymous stealth project, but the evidence remains early and mostly ecosystem-based. Startup Nation Finder identifies the startup as founded in January 2021 by Stas Bank, with a 1-10 employee range, an R&D product stage, undisclosed total funding, and participation in the Genesis Corporate Accelerator and the iHLS Startup Accelerator; the same profile describes FBG sensing for strain, temperature, vibration, and acoustic signatures. The Israel Innovation Authority's company entry identifies the legal entity as PERCIV ADVANCED SENSING LTD, lists Stas Bank as CEO and co-founder, places it in R&D, and records the company's participation in the Applied Research Consortia / Applied Research Fund with a 2025 winning year and additional funding. The Authority page gives an established year of 2019, while the Finder and Israeli corporate-registration references point to a 2021 founding or incorporation timeline. That discrepancy is recorded rather than silently resolved: 2019 may reflect an earlier project or activity date, whereas 2021 is the more consistently reported startup and legal-entity date. PerCiv's own current website strengthens the product evidence by naming the membrane-monitoring problem, describing the non-invasive method as patented, and identifying specific operating decisions the system is designed to support. LinkedIn provides a current company description, a 2-10 employee range, and the claim that membrane maintenance generates $10B in annual losses, but that loss estimate is company-reported and not independently audited here. No priced venture round, institutional investor, patent grant, customer reference, independent benchmark, regulatory certification, or production deployment is disclosed. The 2025 public research support is meaningful third-party validation of technical relevance, not proof of product-market fit or commercial scale.

**Founders and team background.** The public record names Stas Bank as PerCiv's CEO and co-founder, founder, and principal management contact. The corporate and ecosystem pages do not provide a full executive biography, prior exits, academic credentials, or a complete technical-team roster. LinkedIn publicly associates Ziv Glasser with PerCiv and reports four visible employees, while the company-size band is 2-10. That is enough to establish a real operating team but not enough to make claims about deep membrane-process experience, optical engineering depth, software leadership, or commercial selling capability. The founder's continued appearance as the company contact and the company's progression from accelerator participation to a 2025 Israel Innovation Authority applied-research award are constructive signals for founder continuity and technical persistence. The missing information is equally important: there is no public evidence of a named desalination operator on the team, a water-utility procurement veteran, a patent attorney or IP history, a manufacturing partner, or a senior enterprise-sales hire. For this product category, team diligence should focus on whether the founders can bridge several disciplines that are each difficult in isolation: FBG and optical instrumentation, membrane chemistry and fluid dynamics, signal processing, machine-learning validation, industrial networking, and the slow risk-managed buying process of water and process industries. The early-stage company may be deliberately keeping its staff and method confidential, but a customer-ready deployment will require support, installation, calibration, and model-monitoring capability beyond a small research group. Public bios, technical papers, patent applications, reference installations, and evidence of hiring in these disciplines would materially improve confidence in execution capacity.

**Competitive dynamics and defensibility.** PerCiv competes against both technology vendors and the status quo. **DuPont Water Solutions** (large membrane supplier with process expertise and installed-base access) and **Toray Industries** (global membrane manufacturer with engineering and service relationships) can embed diagnostics into broader membrane offerings. **Veolia Water Technologies** (industrial water and desalination integrator with plant-operation and digital-service capabilities) and **SUEZ Water Technologies & Solutions** (water-treatment systems, services, and monitoring) compete through bundled projects where a standalone sensor may be difficult to displace. **Luna Innovations** (fiber-optic sensing and distributed monitoring) represents an adjacent instrumentation supplier that could offer competing optical hardware or become a partner. The most persistent incumbent is **scheduled maintenance plus SCADA, manual sampling, pressure-trend analysis, and periodic membrane autopsy**: it is imperfect but familiar, budgeted, and already integrated into plant operations. PerCiv's potential edge is not simply using fiber optics; it is a proprietary placement and interpretation method designed to extract actionable condition data from existing spiral membrane assemblies without process intrusion. A successful system could create a data advantage from repeated membrane histories, generate switching costs through plant integration, and deliver measurable savings through fewer premature replacements, better cleaning timing, reduced energy waste, and earlier leak detection. Those advantages are not yet established. A durable moat would require demonstrated accuracy across diverse sites, a meaningful library of labeled failure modes, validated patents that cover the method rather than a narrow implementation, low installation cost, and enough integration depth that membrane suppliers or plant operators treat PerCiv as infrastructure rather than an optional dashboard.

**Defense, security, and resilience dual-use relevance.** PerCiv's dual-use case is strongest as water and industrial resilience infrastructure, not as a disclosed defense product. Military bases, defense-production sites, semiconductor fabs, power stations, hospitals, food plants, and municipal water systems all depend on filtration assets whose failure can interrupt operations or reduce access to safe and usable water. A non-invasive method that detects degradation, verifies cleaning, finds leaks, and helps operators avoid wasteful pressure increases could improve continuity in facilities that may have limited maintenance access, constrained spare parts, or a high cost of shutdown. The same technology could be relevant to forward or remote facilities if it can tolerate harsh conditions, operate with intermittent connectivity, and provide reliable local alarms. Its semiconductor and power-industry applicability also gives it strategic relevance beyond water utilities because trusted production of chips and electricity depends on stable process water and filtration. These are credible commercial and resilience applications of the same core sensing system, so dual_use is set to true. The calibration is essential: no public source reviewed here establishes a defense customer, Israeli Ministry of Defense contract, military trial, classified deployment, MIL-STD qualification, cyber-secure plant integration, or operation in a contested environment. PerCiv should therefore be treated as a resilience-enabling industrial technology with a plausible route into security-sensitive infrastructure, not as a fielded defense capability. The next evidence threshold would be a named critical-infrastructure or defense-industrial pilot, documented performance during abnormal operations, secure integration controls, and proof that the system can keep producing useful diagnostics when SCADA connectivity or cloud access is degraded.

**Growth stage, trajectory, and key diligence risks.** PerCiv is appropriately classified as early: its founding or incorporation is recent, its public product is still described in R&D terms, its team is small, funding is undisclosed apart from public research support, and commercial traction is not referenceable. The trajectory is encouraging in one narrow sense: the company has moved from accelerator participation and an ecosystem profile to a sharper current product message, a stated patented monitoring method, and Israel Innovation Authority applied-research support in 2025. It remains too early to infer recurring revenue, repeatable installation economics, or a scalable sales motion. The highest-priority diligence points are: (1) **technical proof**, including blinded detection results, membrane-level localization accuracy, performance by membrane chemistry and feedwater, and comparison with existing pressure and conductivity analytics; (2) **deployment economics**, including sensor and interrogator bill of materials, installation time, recalibration burden, software pricing, and quantified energy or membrane-replacement savings; (3) **IP and ownership**, including the scope and status of the claimed patent, inventorship, freedom to operate, and whether the company controls the core optical design; (4) **commercial validation**, including paid pilots, customer references, conversion rates, and the role of EPCs or membrane suppliers; (5) **team completeness**, particularly membrane-process engineering, optical hardware, industrial software, and field service; (6) **resilience and cybersecurity**, including edge operation, data ownership, segmentation, and secure remote updates; and (7) **financing risk**, since a hardware-plus-industrial-sales company may need more runway than accelerator and grant support provide. PerCiv is therefore a high-potential, evidence-limited strategic watch candidate whose value depends on proving that its claimed visibility is accurate enough to change plant decisions.