GOTEAM

**Product and problem.** GOTEAM addresses the detection and identification gap created by small, low, autonomous, radio-silent, or fiber-optic-controlled drones. The company positions O-Lifeguard as a passive AI thermal counter-UAS layer that can be mounted on a vehicle, tripod, rooftop, or other fixed site, then feed an operator or an existing command-and-control system with a track rather than attempting to jam or physically defeat the aircraft. This is a meaningful problem in force protection and civilian airspace security: RF sensors depend on a detectable control link, acoustic systems depend on a sufficiently loud rotor signature, and radar can struggle with small low-altitude objects or can be difficult to deploy around crowded spectrum and civilian environments. GOTEAM says its system is deployable in hours, operates day and night, and can work offline with Ethernet, cellular, or satellite backhaul when available. The public product positioning is detection and decision support, not a complete defeat chain, so the value proposition depends on improving warning time and cueing a customer’s existing mitigation layer.

**Core technology and operation.** The published architecture is unusually specific for an early-stage company. GOTEAM describes an uncooled long-wave infrared core that streams full 16-bit frames at 30-plus frames per second, with EO/IR options and weather-rated optics. Its processing pipeline first applies morphological filtering to reduce slowly changing structures such as clouds, sky gradients, and warm horizons. A purpose-trained neural network then proposes drone candidates from the processed frame, including targets represented by only a small number of pixels. A frame-to-frame tracker links candidates by position and velocity and applies a persistence gate so a single hot pixel or transient clutter does not become a confirmed track. The system then produces calibrated distance, azimuth, and elevation and sends an annotated clip and metadata to the GOTEAM Hub or an external C2 system. The company claims an edge latency below 20 milliseconds, detection up to 400 meters, NVIDIA Jetson processing, an IP67 sealed unit, and operation from -20 C to +55 C. These are vendor specifications and should be tested against range, weather, aspect, clutter, and false-alarm conditions rather than treated as independently certified performance.

**Market, customers, and go-to-market.** GOTEAM is selling into a broad but procurement-intensive market: military bases and field camps, armor and convoys, airports, police and public events, prison perimeters, VIP protection, energy and water sites, ports, dams, defense plants, and government facilities. The system’s passive character gives it a plausible entry point where operators cannot add a transmitter, obtain a spectrum license quickly, or use indiscriminate jamming near people and aircraft. Its open-architecture message is aimed at integrators already operating RF, acoustic, radar, or effectors, making GOTEAM a partner sensor rather than a replacement for the full C-UAS stack. The careers page identifies sales roles for the United States, APAC, and Israel/EMEA, and specifically references defense, homeland-security, airport, energy, integrator, and end-user relationships. That suggests a channel-led international expansion model with direct technical support during installation and calibration. The same page advertises OEM and algorithm roles, implying that sensor integration, SDK licensing, or embedded deployment may become additional routes to market. No named customer, purchase order, recurring revenue figure, or conversion rate is publicly disclosed.

**Traction, funding, and third-party validation.** Public evidence shows a live product site, a technical explanation of the detection pipeline, a fleet-management concept, a public FlyBox evaluation sandbox, and an active recruiting effort. GOTEAM says its models have been trained on more than 500,000 real-world thermal detections from live deployments and that its neural network uses tens of thousands of hand-verified thermal drone frames. It also presents live-output demonstration clips and says uploaded customer footage can be processed on a real Field device, producing an annotated video. These are useful signs of productization and data operations, but they remain company-reported evidence; the reviewed public materials do not identify the deployments, independent test protocol, precision and recall, false-alarm rate, contract value, or customer reference. The Hub article describes 30-day retention of annotated video and metadata, raw-frame reprocessing, sensor-health history, and fleet-wide model rollout in about ten seconds, which indicates attention to operational lifecycle rather than a one-off laboratory demo. An Israeli corporate-information record identifies GOTEAM LTD as an active private company incorporated on January 6, 2019, with a Haifa address and Baruch Glick as a director. No public funding round, valuation, institutional investor, grant amount, or audited financial result was found in the reviewed sources, so funding stage and commercial scale remain open diligence items.

**Founders and team.** The public company material identifies Baruch Glick as founder and author of the technical blog posts. Glick writes that he spent decades building thermal-imaging detection at SCD before founding GOTEAM; the legal-company record separately identifies him as a director. That background is strategically relevant because the product’s hardest stated problem is not merely acquiring a thermal camera but converting weak thermal returns into reliable field detections under clutter and then maintaining the resulting algorithm across a fleet. The careers page characterizes GOTEAM as a small, senior team with a decade of autonomous-flight engineering and is recruiting for EO/IR algorithms, mechanical engineering, support engineering, and regional defense sales. Those roles map coherently to the product’s needs: signal processing and model optimization, ruggedized mounts and enclosures, field calibration, and government or prime-integrator procurement. However, the public record does not provide a complete executive roster, employee count, named technical leads, patent portfolio, academic affiliations, or independently verified prior programs. The team thesis is therefore strong at the founder-domain level but not yet a complete assessment of organizational depth or delivery capacity.

**Competitive dynamics and edge.** GOTEAM competes in a crowded layered C-UAS market, but it is not making the same claim as an RF-cyber takeover provider or a radar-first system. Its proposed edge is a compact, passive, fixed-stare EO/IR layer that can see the aircraft itself, including a target that does not transmit, while running classification and tracking at the sensor. The company argues that a fixed sensor can inspect the full scene every frame and preserve multiple tracks, whereas a single PTZ camera must choose a sector and often needs an external cue. It also claims that SAPIENT and BSI Flex 335 v2 C2 output makes the system easier to fuse with existing sensors and effectors. The strongest potential moat is not the commodity thermal core; it is the accumulated field dataset, failure analysis, model update workflow, and integration reliability around the core. The counterargument is equally important: established companies already sell passive EO/IR, radar, RF, acoustic, and sensor-fusion systems, and customers may prefer a single prime with validated end-to-end responsibility. GOTEAM must prove that its lower-power and lower-complexity architecture produces acceptable detection probability and false-alarm behavior across weather, terrain, swarms, birds, and adversarial tactics.

**Defense, security, and resilience relevance.** Dual-use relevance is direct because the same core capability supports both military force protection and civilian protection of airports, utilities, ports, water infrastructure, public events, prisons, and VIP movements. A passive sensor that emits no RF signal can preserve operational discretion for a convoy or forward base and can avoid adding interference around an airport or dense urban event. Detection of autonomous or fiber-optic drones is also strategically relevant because it targets a physical aircraft signature rather than assuming that the adversary exposes a radio-control link. The system’s open API and claimed SAPIENT compatibility could allow it to cue an existing interceptor, jammer, or operator workflow without owning the effectors. That makes GOTEAM a potentially useful allied supply-chain component rather than a complete weapons system. The limits matter: the company publicly says it does not jam, intercept, or shoot down drones, and thermal detection can be constrained by range, weather, obscuration, background temperature, target aspect, and line of sight. There is no public evidence in the reviewed material of an IDF fielding, a named foreign military contract, a certified government program, or an operational kill chain. Its security value is therefore credible at the sensing and resilience layer, while fielded defense effectiveness remains to be verified.

**Stage, trajectory, and diligence risks.** GOTEAM should be classified as early commercial or early scale-up for this database. The legal entity dates to 2019, but the public O-Lifeguard and Hub positioning, current technical blog, FlyBox workflow, and geographically distributed sales hiring are visible in 2026; that combination suggests a company moving from engineering and demonstrations toward repeatable deployments, not a mature public supplier. The trajectory could be attractive if the company converts its claimed field data into independently benchmarked performance, wins reference customers, and becomes the passive optical layer inside larger C-UAS and critical-infrastructure programs. Key diligence points are: (1) verify the 400-meter range, sub-20-millisecond latency, 500,000-plus detection dataset, and false-alarm metrics under representative conditions; (2) identify the deployments behind the company’s “live” and “field” language; (3) understand unit economics, manufacturing partners, export classification, and support obligations; (4) test the Hub’s cybersecurity, offline behavior, model-signing, update rollback, and data-retention controls; (5) clarify patent ownership and freedom to operate around sensor processing and tracking; and (6) assess whether a small team can satisfy defense qualification, procurement, and multi-country support requirements. These unknowns warrant a medium risk rating and keep the strategic assessment positive but explicitly provisional.