Peregrine.ai
AI-powered perception systems for smart cameras, delivering contextual awareness in mobility.
Website: https://peregrine.ai/
Cover Block
From the public record
| Field | Detail |
|---|---|
| Name | Peregrine.ai |
| Tagline | AI-powered perception systems for smart cameras, delivering contextual awareness in mobility [Peregrine.ai, retrieved 2026] |
| Headquarters | Berlin, Germany |
| Founded | 2019 |
| Stage | Series D+ |
| Business model | SaaS |
| Industry | Logistics / Supply Chain |
| Technology | AI / Machine Learning |
| Geography | Western Europe |
| Growth profile | Venture Scale |
| Founding team | Co-Founders (3+) |
| Funding label | $100M+ |
| Total disclosed funding | ~$470,000,000 [Crunchbase, retrieved 2026]; [Peregrine, May 2024]; [Peregrine newsroom]; [Peregrine, June 2026] |
Links
From the public record
- Website: https://peregrine.ai/
- LinkedIn: https://www.linkedin.com/in/najavonschmude/
The Short Version
PUBLIC Peregrine.ai develops AI-powered perception software that turns cameras into real-time contextual awareness systems for mobility use cases, and it merits attention now because the company has moved from early product claims to a broader commercial posture that includes telematics partnerships, a disclosed Series D+, and evidence of active hiring into go-to-market and engineering functions [Peregrine.ai, retrieved 2026] [Peregrine.ai, Sept 2025] [Peregrine, June 2026] [Peregrine Greenhouse careers, October 2026]. The public record ties the company to Berlin, a 2019 founding date, and a three-person founding group that includes Naja von Schmude, Steffen Heinrich, and Jorit J. K. Schmelzle, with multiple public profiles identifying von Schmude as co-founder and CTO and Schmelzle as co-founder and CPO [Crunchbase, retrieved 2026] [LinkedIn, retrieved 2026] [me.sh, retrieved 2026].
The product case rests on edge inference rather than a fleet-facing application sold directly: Peregrine says its software runs on-device, processes camera frames to action in under 85 milliseconds, requires no new hardware, and is sold as an SDK to telematics providers and platforms rather than to fleets themselves [Peregrine.ai, retrieved 2026]. That architecture, if it holds in production, could matter commercially because the company is positioning privacy, bandwidth efficiency, and alert quality as decision criteria, with claims of anonymization before storage, 95 percent less bandwidth than video streaming in selected deployments, and more than 90 percent fewer false alerts in selected deployments, although those performance figures are company-supplied and should be treated accordingly [Peregrine.ai, retrieved 2026].
Go-to-market evidence is still mostly partnership-led in the public materials. The clearest named examples are collaborations with Webfleet, Bridgestone's fleet management solution, and Linqo GmbH, alongside a flagship product called Peregrine Vision for risk detection, driver behavior analysis, and event-based alerts [Peregrine.ai, Sept 2025] [Peregrine.ai, May 2025].
Capital formation has been substantial relative to the maturity of the disclosed product footprint. Structured sources point to an undisclosed seed in 2021, a separate seed amount of $1 million reported by Vektor Partners, a $30 million Series B in May 2024, a $190 million Series C in March 2025, and a $250 million Series D+ in June 2026, for roughly $471 million disclosed in aggregate based on the reported figures (estimated) [Crunchbase, retrieved 2026] [Vektor Partners] [Peregrine, May 2024] [Peregrine newsroom, retrieved 2026] [Peregrine, June 2026]. The stated model is SaaS, but the commercial surface described publicly looks closer to embedded B2B software distributed through telematics and platform partners, which is an important distinction for diligence on pricing power, channel dependency, and deployment cycle length [Peregrine.ai, retrieved 2026].
Over the next 12 to 18 months, the key watchpoints are straightforward: whether partner-led launches convert into repeatable deployments across Europe and North America, whether the heavy hiring signal translates into execution rather than organizational drag, and whether the company's public traction claims become easier to reconcile across sources [Peregrine.ai, May 2025] [Peregrine.ai, Sept 2025] [Peregrine Greenhouse careers, October 2026]. The upside case is a privacy-aware perception layer that can sit inside existing fleet and camera infrastructure; the analytical constraint is that many of the most attractive product and traction claims remain company-issued rather than independently corroborated [Peregrine.ai, retrieved 2026] [Fortune, June 2026].
Unconfirmed -- Core company identity and financing are partially corroborated by Crunchbase and company materials, but most product performance and traction claims in this section are company-issued with limited independent verification.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Series D+ |
| Business Model | SaaS |
| Industry / Vertical | Logistics / Supply Chain |
| Technology Type | AI / Machine Learning |
| Geography | Western Europe |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding | $100M+ total disclosed, approximately $470,000,000 |
The Company in Brief
PUBLIC
Peregrine.ai presents as a Berlin-based company building AI-powered perception systems for smart cameras, with a product focus on real-time contextual awareness in mobility use cases [Crunchbase, retrieved 2026] [Peregrine.ai, retrieved 2026]. Public records in the provided source set place the company’s founding in 2019, and founder references consistently identify Naja von Schmude, Steffen Heinrich, and Jorit J. K. Schmelzle as members of the founding team, though title detail is stronger for von Schmude and Schmelzle than for Heinrich [Crunchbase, retrieved 2026] [LinkedIn, retrieved 2026].
The operating identity is somewhat split in public materials. Crunchbase lists Peregrine.ai, while LinkedIn profiles and company materials also reference Peregrine Technologies GmbH or Peregrine Technologies, which suggests a branding or entity overlap rather than a contradiction, but the exact legal-entity structure is not established in the source set [Crunchbase, retrieved 2026] [LinkedIn, retrieved 2026] [Peregrine.ai, retrieved 2026].
The milestones that are clearly documented show a company moving from early financing into a later-stage capital profile. Crunchbase records a 2021 seed round led by Vektor Partners, and company announcements then show a $30 million Series B in May 2024, a $190 million Series C referenced in the company newsroom in March 2025, and a $250 million Series D+ in June 2026 led by Fifth Down Capital [Crunchbase, retrieved 2026] [Peregrine, May 2024] [Peregrine newsroom, retrieved 2026] [Peregrine, June 2026]. Over that period, the website and related materials also show the product maturing from a general smart-camera perception stack to named transportation offerings such as Peregrine Vision and partner-led launches with Linqo and Webfleet in 2025 [Peregrine.ai, retrieved 2026] [Peregrine.ai, May 2025] [Peregrine.ai, Sept 2025].
Single-source, plausible -- Company identity, headquarters, founding year, and financing milestones are supported by Crunchbase and company materials, but the exact legal-entity structure is only partially corroborated in the public sources provided.
What They Have Built
MIXED Peregrine.ai presents itself as an edge AI software company rather than a camera hardware vendor. Its core claim is that the software reads scenes from existing smart cameras in real time, on the device, for road safety, driver assistance, and site security, with frame-to-action latency of under 85 milliseconds, no new hardware requirement, and data anonymization before storage according to the company website [Peregrine.ai, retrieved 2026]. The same materials say selected deployments used 95% less bandwidth than video streaming and generated more than 90% fewer false alerts, but those performance figures are company-selected and not independently verified [Peregrine.ai, retrieved 2026].
The commercial product surface is clearer in transportation than in a broader computer vision category. Peregrine Vision, introduced publicly in 2025, is described as a video telematics product that delivers real-time, privacy-compliant insights for risk detection, driver behavior analysis, and event-based alerts [Peregrine.ai, Sept 2025]. The company says it sells this software as an SDK to telematics providers and platforms rather than directly to fleets, and it has publicly named Webfleet and Linqo as partners for visual intelligence and integrated video telematics offerings in Europe [Peregrine.ai, Sept 2025] [Peregrine.ai, May 2025]. Trade coverage broadly aligns with that partner-led positioning, describing the Webfleet launch as a driver assistance system that detects and contextualizes road events as they happen [Fleetworld] [Transport Engineer] [Business Motoring].
Some implementation detail is visible around the edges, but the public record remains thin on the underlying stack. A LinkedIn profile for Jonathan Huber says the company rebuilt Aplomado, a cloud platform that connects live fleet data, video events, GPS traces, risk scores, and analytics for customers managing fleets, which suggests Peregrine is not only shipping on-device inference but also an operational analytics layer around alerts and vehicle data [Jonathan Huber LinkedIn, retrieved 2026]. Hiring adds a modest technical signal: open roles in applied AI, platform engineering, and product security point to continued investment in model deployment, backend infrastructure, and security controls, though any stack conclusion beyond that would be an inference from job postings rather than a disclosed architecture [Greenhouse, October 2026] [Peregrine Greenhouse careers, retrieved 2026] [LinkedIn jobs].
Unconfirmed -- This section relies primarily on company website and press materials, with limited secondary corroboration from trade publications and public job listings.
Market Size and Demand
PUBLIC The market matters now because camera hardware is already deployed at scale, while privacy, bandwidth, and driver-safety requirements are pushing operators toward on-device intelligence rather than raw video streaming [Peregrine.ai, retrieved 2026] [Peregrine.ai, Sept 2025].
The available public record does not support a clean TAM, SAM, and SOM build for Peregrine.ai from named third-party market studies, so the market has to be framed through the company’s stated use cases and buying motion rather than a synthetic sizing model. On the evidence in hand, the addressable wedge is video telematics and smart-camera software for mobility, especially software sold through telematics providers and platform partners instead of directly to fleets [Peregrine.ai, retrieved 2026] [Peregrine.ai, Sept 2025]. That matters because it places Peregrine in a software layer that can ride existing camera installs and fleet-management relationships, assuming partners choose to embed the SDK rather than build comparable perception capabilities internally [Peregrine.ai, retrieved 2026].
Demand signals in the source set are concrete, if still mostly company-supplied. Peregrine says its systems read scenes in real time on the device, act in under 85 milliseconds, require no new hardware, anonymize data before storage, and in selected deployments use 95% less bandwidth than streaming video while reducing false alerts by more than 90% [Peregrine.ai, retrieved 2026]. Those claims point to the main buyer pain points in this category: bandwidth cost, driver-risk detection, privacy compliance, and alert fatigue. The partnership pattern is also telling. The Webfleet collaboration and the Linqo launch suggest demand is being intermediated through established fleet software and telematics channels, which is often how European mobility software reaches commercial fleets at scale [Peregrine.ai, Sept 2025] [Peregrine.ai, May 2025].
| Market lens | Public evidence | Read-through for demand |
|---|---|---|
| Installed-camera upgrade path | Peregrine says its software requires no new hardware and runs on-device [Peregrine.ai, retrieved 2026] | Lowers adoption friction if current camera fleets can be upgraded through software rather than vehicle retrofits |
| Privacy-first video intelligence | Peregrine says data is anonymized before storage and positions Peregrine Vision as privacy-compliant [Peregrine.ai, retrieved 2026] [Peregrine, Sept 2025] | Supports adoption in Europe and regulated fleet contexts where continuous video recording is sensitive |
| Channel-led distribution | Peregrine sells an SDK to telematics providers and platforms, and has named partnerships with Webfleet and Linqo [Peregrine.ai, retrieved 2026] [Peregrine.ai, Sept 2025] [Peregrine.ai, May 2025] | Suggests the near-term market is as much platform enablement as direct fleet software spend |
| Measured expansion signal | Peregrine’s careers page says it operates across 30+ states and two countries [Peregrine careers page, retrieved 2026] | Indicates the product is being positioned for cross-border deployment, though not yet broad global coverage |
The table points to a market that is less about selling another dashboard and more about upgrading existing telematics stacks with machine perception. That can be an efficient position if partners treat contextual video intelligence as additive rather than core to their own roadmap.
Adjacent markets are broad enough to matter, but they should be separated carefully. The direct adjacency is video telematics for fleets, where the product claims center on risk detection, driver behavior analysis, and event-based alerts [Peregrine, Sept 2025]. A second adjacency is road-safety and driver-assistance software, especially where on-device inference can supplement or improve traditional dashcam workflows [Fleetworld] [Transport Engineer] [Business Motoring]. Site security is also mentioned on the company site, but the public evidence here is thinner than for transportation, so it reads more like an extensible product surface than a confirmed commercial beachhead [Peregrine.ai, retrieved 2026].
Regulation and macro conditions appear to favor this category, even if the section’s source base is limited. Privacy-sensitive environments in Europe reward architectures that minimize stored personal data, while fleet operators remain under pressure to control insurance, safety, and operating costs. Against that backdrop, on-device perception has a straightforward economic argument if the cited bandwidth and false-alert improvements prove repeatable beyond selected deployments [Peregrine.ai, retrieved 2026]. The main caveat is evidentiary rather than conceptual: the strongest market claims in the file come from the company and partner announcements, so the category logic is credible, but the size and pace of spend still need more independent validation.
Unconfirmed -- This section relies primarily on company and partner materials, with limited independent corroboration for market structure and no named third-party TAM study in the source set.
Who Else Is Fighting for This
Competitive Positioning
MIXED Peregrine.ai is positioned less like a full-stack fleet software vendor and more like an infrastructure layer for camera-based intelligence, which puts it in competition with direct video telematics entrants on one side and broader platform incumbents on the other [Peregrine.ai, retrieved 2026] [Peregrine, Sept 2025].
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Peregrine.ai | AI perception software and SDK for smart cameras and video telematics partners | Series D+; total disclosed funding of about $470M | On-device scene understanding, software sold via SDK to telematics providers rather than directly to fleets | [Crunchbase, retrieved 2026] [Peregrine.ai, retrieved 2026] [Peregrine, June 2026] |
| Rarestep | Named competitor in the structured research | Not established in provided sources | Not established in provided sources | [Structured facts] |
The competitive map is fragmented because Peregrine sits across at least two buying motions. In transportation, the company is presenting Peregrine Vision as a privacy-oriented visual intelligence layer for telematics platforms, with Webfleet and Linqo cited as go-to-market partners rather than end customers [Peregrine, Sept 2025] [Peregrine, May 2025]. That suggests one set of rivals will be other perception or video analytics vendors trying to supply the same platforms, while another set will be telematics incumbents that prefer to build computer vision features internally once demand is proven.
The company has a credible edge in distribution efficiency, but the evidence points to an edge that is useful rather than permanent. Peregrine says its software runs on-device, requires no new hardware, and uses less bandwidth than continuous video streaming in selected deployments, all of which matter when the buyer is a telematics platform trying to upgrade an installed base without forcing a hardware refresh [Peregrine.ai, retrieved 2026]. The Webfleet collaboration also matters because it indicates Peregrine can access established fleet channels indirectly, which is often faster than selling fleet-by-fleet from scratch [Peregrine, Sept 2025] [Destination Net Zero Magazine] [Fleetworld]. That said, most of these differentiators are company-described product claims, and SDK position can be perishable if larger channel partners decide the margin pool belongs inside their own stack.
Exposure is easiest to see at the channel layer. Because Peregrine says it sells to telematics providers and platforms rather than directly to fleets, it does not appear to control the end-customer relationship in its transportation motion [Peregrine.ai, retrieved 2026]. A competitor such as Rarestep, if it owns a more direct route into fleet operators or bundles hardware, software, and workflow into one contract, could have an advantage in sales simplicity even if its perception layer is less modular. Peregrine is also exposed to adjacent substitutes from existing camera, telematics, or fleet management vendors that can frame visual intelligence as a feature extension rather than a new procurement category.
Over the next 18 months, the most plausible competitive scenario is a sorting between infrastructure suppliers and integrated operators. Peregrine is a winner if telematics platforms continue to prefer neutral SDK partners that can add driver assistance, risk detection, and event understanding quickly across existing camera estates [Peregrine.ai, retrieved 2026] [Peregrine, Sept 2025]. Rarestep is a potential winner if buyers shift toward bundled solutions with fewer integration steps, especially in cost-sensitive fleets that value one throat to choke over configurability. The likely loser, if platform owners bring perception in-house, is any independent vendor whose differentiation rests mainly on model performance claims that are difficult for customers to verify independently.
Opportunity
PUBLIC
The prize here is straightforward: if Peregrine.ai becomes the default perception layer inside commercial camera and telematics stacks, it could sit in the workflow between raw video and operational decision-making across a very large installed base of mobility and safety systems [Peregrine.ai, retrieved 2026] [Peregrine.ai, Sept 2025].
The headline opportunity is not simply to sell another camera feature set. It is to become the software layer that lets existing cameras produce real-time, privacy-aware, on-device decisions without forcing customers to replace hardware, a proposition the company states can cut bandwidth use and reduce false alerts in selected deployments [Peregrine.ai, retrieved 2026]. That outcome looks reachable, not merely aspirational, because the company is already positioning itself as an SDK supplier to telematics providers rather than a direct fleet seller, which is the more scalable insertion point if the goal is broad embedded distribution [Peregrine.ai, retrieved 2026]. The two public partnership announcements that matter most in that context are Webfleet, Bridgestone's fleet management platform, and Linqo in the European fleet market, both of which suggest Peregrine is testing channel-based expansion rather than relying on one-by-one direct sales [Peregrine.ai, Sept 2025] [Peregrine.ai, May 2025].
The upside paths are easier to see when separated into a few concrete routes.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Embedded telematics standard | Peregrine becomes a preferred perception SDK inside established fleet and telematics platforms, with partners distributing the product into their installed bases | The Webfleet visual intelligence launch and the Linqo integration give Peregrine two public reference points for platform-led adoption [Peregrine.ai, Sept 2025] [Peregrine.ai, May 2025] | The company says it sells to telematics providers and platforms rather than directly to fleets, which is consistent with an infrastructure model that can scale through partners [Peregrine.ai, retrieved 2026] |
| Edge AI upgrade cycle | Existing camera deployments adopt on-device contextual awareness because Peregrine says the software requires no new hardware and processes events in under 85 ms on device [Peregrine.ai, retrieved 2026] | A broader customer preference for bandwidth savings, lower alert fatigue, and privacy-sensitive processing could make retrofit software materially easier to buy than full hardware replacement [Peregrine.ai, retrieved 2026] | The company frames its core value as real-time scene understanding on device, with anonymization before storage and lower bandwidth use in selected deployments, all of which address practical procurement constraints [Peregrine.ai, retrieved 2026] |
| Multi-product mobility intelligence platform | Peregrine extends from video telematics into a broader operating layer for fleet risk, driver assistance, and analytics workflows | Peregrine Vision formalizes a flagship product, while Aplomado is described by a company employee as the cloud platform connecting live fleet data, video events, GPS traces, risk scores, and analytics [Peregrine, Sept 2025] [Jonathan Huber LinkedIn, retrieved 2026] | The available product surface already spans on-device perception plus cloud-side workflow and analytics, which is the architecture usually required for expansion beyond a single feature [Peregrine.ai, retrieved 2026] [Jonathan Huber LinkedIn, retrieved 2026] |
What compounding looks like, if this works, is a channel-and-data flywheel rather than a classic consumer network effect. Each platform integration can put the software in front of many downstream fleets, each additional deployment can generate more edge-case footage and operational feedback, and those learnings can improve alert quality, which in turn makes the product easier for partners to resell and for end customers to keep switched on [Peregrine.ai, retrieved 2026] [Peregrine.ai, Sept 2025]. The signs of early compounding are still company-reported, but they are directionally relevant: Peregrine says it has partnerships in Europe and with a major fleet platform, while third-party trade coverage repeats the Webfleet launch as a next-generation driver assistance product rather than a one-off pilot [Fleetworld] [Business Motoring] [Transport Engineer]. The hiring footprint also matters at the margin. Ninety-three open roles as of October 2026, including applied AI, platform engineering, product security, customer roles, and pricing, point to a business investing across product, go-to-market, and implementation at the same time [Peregrine careers page, retrieved 2026] [Greenhouse, October 2026] [LinkedIn jobs].
The size of the win is harder to anchor because the public comps in the supplied record are thin, but one valuation datapoint is concrete. Peregrine's June 2026 financing announcement states a $6.8 billion valuation alongside a $250 million Series D+, with Fortune separately reporting the same financing event [Peregrine, June 2026] [Fortune, June 2026]. On that basis, a credible upside case is not whether the company can become venture-scale, but whether it can grow into a much larger infrastructure position inside mobility software. If the embedded telematics standard scenario plays out and the company converts partner distribution into a category-level software layer, an outcome above its last private valuation is conceivable, though any figure beyond $6.8 billion would be a scenario, not a forecast [Peregrine, June 2026] [Fortune, June 2026]. The evidence in the public record supports taking that upside seriously, even if most operating claims still rest on company disclosures.
Unconfirmed -- This section relies materially on company announcements, company website product claims, and trade re-reporting with limited independent verification beyond Fortune's coverage of the June 2026 financing.
Sources
From the public record
[Peregrine.ai, retrieved 2026] AI-powered vision, for smarter cameras | https://peregrine.ai/
[Crunchbase, retrieved 2026] Peregrine.ai - Crunchbase Company Profile & Funding | https://www.crunchbase.com/organization/peregrine-ai
[Peregrine, May 2024] Peregrine Announces $30M Series B to Accelerate Growth with Government & Public Safety Agencies | https://peregrine.io/resources/peregrine-announces-30m-series-b-to-accelerate-growth-with-government-public-safety-agencies
[Peregrine newsroom, retrieved 2026] Newsroom | https://peregrine.io/newsroom
[Peregrine, June 2026] Peregrine Technologies Raises $250 Million Series D at $6.8 Billion Valuation | https://peregrine.io/resources/peregrine-technologies-raises-250-million-series-d-at-68-billion-valuation
[Peregrine Greenhouse careers, October 2026] Careers | https://job-boards.greenhouse.io/peregrinetechnologies?error=true
[LinkedIn, retrieved 2026] Naja von Schmude - Peregrine Technologies GmbH | LinkedIn | https://www.linkedin.com/in/najavonschmude/
[me.sh, retrieved 2026] Jorit J. K. Schmelzle | https://me.sh/
[Peregrine.ai, Sept 2025] Webfleet and Peregrine.ai collaborate on Visual Intelligence solution | https://peregrine.ai/webfleet-and-peregrine-ai-collaborate-on-visual-intelligence-solution/
[Vektor Partners, retrieved 2026] Vektor Partners | https://vektor.vc/
[Peregrine.ai, May 2025] Peregrine.ai and Linqo Launch AI-Powered Video Telematics Solution for European Fleet Market | https://peregrine.ai/peregrine-ai-and-linqo-partnership-press-release/
[Fortune, June 2026] Exclusive: The AI company powering public safety operations for the 2026 World Cup just raised $250 million | https://fortune.com/2026/06/22/exclusive-peregrine-nick-noone-ai-public-safety-palantir-2026-world-cup-just-sequoia-capital/
[Fleetworld, retrieved 2026] Webfleet and Peregrine.ai partner on visual intelligence solution | https://fleetworld.co.uk/webfleet-and-peregrine-ai-partner-on-visual-intelligence-solution/
[Transport Engineer, retrieved 2026] Webfleet and Peregrine.ai launch next-generation driver assistance solution | https://transportengineer.org.uk/transport-engineer-news/webfleet-and-peregrine-ai-launch-next-generation-driver-assistance-solution/257725/
[Business Motoring, retrieved 2026] Webfleet and Peregrine.ai partner to launch driver visual intelligence system | https://businessmotoring.co.uk/webfleet-and-peregrine-ai-partner-to-launch-driver-visual-intelligence-system/
[Jonathan Huber LinkedIn, retrieved 2026] Jonathan Huber | LinkedIn | https://www.linkedin.com/
[Greenhouse, October 2026] Staff Software Engineer, Applied AI | https://job-boards.greenhouse.io/peregrinetechnologies/jobs/4740981005
[LinkedIn jobs, retrieved 2026] Peregrine Technologies Jobs | LinkedIn | https://www.linkedin.com/company/peregrine-technologies/jobs
[Destination Net Zero Magazine, retrieved 2026] Webfleet and Peregrine.ai launch next-generation driver assistance solution | https://destinationnetzeromagazine.com/webfleet-and-peregrine-ai-launch-next-generation-driver-assistance-solution/
Articles about Peregrine.ai
- Peregrine.ai’s Perception Systems Run on the Camera Itself — The Berlin startup, backed by Sequoia and Fifth Down, sells an SDK that processes video in 85 milliseconds without streaming to the cloud.