Drogue

Harnessing the power of the ocean to generate affordable, off-grid energy for data centers.

Cover Block

Publicly reported

Field	Value
Name	Drogue
Tagline	Harnessing the power of the ocean to generate affordable, off-grid energy for data centers. [drogue.com]
Headquarters	Washington, D.C., United States [Sam Sipe LinkedIn, retrieved 2026]
Founded	2026 [TechCrunch, November 2026]
Stage	Seed [TechCrunch, November 2026]
Business Model	Hardware + Software
Industry	Deeptech
Technology	Hardware
Geography	North America
Growth Profile	Venture Scale
Founding Team	Solo Founder, Sam Sipe [Jingletree]
Funding Label	$10M+
Total Disclosed Funding	~$20,000,000 [TechCrunch, November 2026]

Summary and Signal

PUBLIC Drogue is a Washington, D.C.-based startup building floating offshore data centers that aim to use ocean thermal energy conversion for both on-board power generation and cooling, a concept that merits attention because it targets two binding constraints in AI infrastructure at once, power availability and heat management [TechCrunch, November 2026]. The timing is unusually early but concrete enough to matter: public reporting says the company was founded in 2026, surfaced roughly six months before its November 2026 debut, and has already outlined a plan to put a multi-megawatt prototype on a conventional ocean-going ship in early 2028 [TechCrunch, November 2026] [MLQ, November 2026].

The core bet is not generic data center capacity but a ship-based platform that colocates computing with offshore energy generation and seawater cooling, which, according to available coverage, could sidestep some land-based interconnection and construction bottlenecks if the engineering works as described [TechCrunch, November 2026] [Jingletree]. Public detail remains thin where investors will care most, including deployment site, customer commitments, vessel economics, and exact power output per ship, so the current case rests more on technical ambition and category timing than on commercial proof points [MLQ, November 2026] [Proofstory, November 2026].

Founder-market fit is directionally interesting. Sam Sipe is identified as co-founder and CEO, with prior engineering leadership at Scale AI and earlier service flying E-2D Hawkeye aircraft from U.S. Navy aircraft carriers, a combination that plausibly maps to mission-critical systems and infrastructure execution, though no additional co-founder or technical founder has been clearly named in public reporting [Linear] [MLQ, November 2026] [Proofstory, November 2026].

Funding gives the company more room than most concept-stage deeptech efforts. TechCrunch reported a $20 million November 2026 seed round, with investors including XYZ Venture Capital, Toyota Ventures, Tamarack Global, Dream, FDVC, Garuda Ventures, Flexcap, Boost VC, and Slow Ventures, and the stated business model is hardware plus software rather than pure project development [TechCrunch, November 2026]. Over the next 12 to 18 months, the key markers to watch are whether Drogue converts design claims into disclosed engineering milestones, expands the technical bench beyond a solo publicly identified founder, and shows evidence of commercial demand or deployment partnerships ahead of the planned 2028 prototype window [Ashbyhq, November 2026] [Ground Floor Careers, August 2026] [MLQ, November 2026].

One source, partially checked -- This section relies primarily on TechCrunch and MLQ, with partial corroboration from careers listings, Linear, and Proofstory; several core operating claims remain company-originated or lightly verified in public reporting.

Taxonomy Snapshot

Axis	Value
Stage	Seed
Business Model	Hardware + Software
Industry / Vertical	Deeptech
Technology Type	Hardware
Geography	North America
Growth Profile	Venture Scale
Founding Team	Solo Founder
Funding	$10M+ total disclosed, approximately $20,000,000 [TechCrunch, November 2026]

Company Overview

PUBLIC

Drogue entered the public record in 2026 with a proposition that is unusually ambitious even by deeptech standards: put data-center capacity at sea and use the ocean itself as both the energy source and cooling system [drogue.com] [TechCrunch, November 2026]. The company is based in Washington, D.C., and the available public materials identify Sam Sipe as co-founder and CEO [drogue.com] [Jingletree].

The chronology is short but clear on the points that matter. Public reporting says the company was founded roughly six months before its November 2026 debut, and that same month it disclosed a $20 million seed round, giving it capital to move from concept framing toward a first prototype plan [TechCrunch, November 2026]. The website and launch coverage consistently describe the initial milestone as a multi-megawatt shipboard prototype targeted for early 2028, which places Drogue today in a pre-deployment, prototype-building phase rather than commercial rollout [drogue.com] [TechCrunch, November 2026].

One source, partially checked -- This section relies primarily on the company website, with chronology and funding timing corroborated by TechCrunch.

The Product and the Stack

MIXED

The technical claim is straightforward to state and harder to underwrite at this stage: Drogue says it is building floating offshore data centers that use ocean thermal energy conversion, or OTEC, to produce power while using seawater for cooling [TechCrunch, November 2026] [drogue.com]. Public coverage describes the platform as ship-based and aimed at energy-intensive computing, including AI workloads, with the company positioning offshore deployment as a way to colocate compute, power generation, and heat rejection in one system [TechCrunch, November 2026] [Jingletree]. That is an unusual architecture, but the evidence in public remains early, consisting mainly of launch coverage and company materials rather than an observed deployment or third-party technical validation [TechCrunch, November 2026] [MLQ, November 2026].

The disclosed build status is still pre-deployment. TechCrunch and MLQ both report that Drogue has designed a multi-megawatt prototype and plans to place its first prototype on a conventional ocean-going ship in early 2028 [TechCrunch, November 2026] [MLQ, November 2026]. The hiring footprint gives a narrow but useful read on the engineering stack: thermodynamics, electrical and controls, and simulation appear to be priority functions [Ashbyhq, November 2026] [Ground Floor Careers, August 2026]. A simulation-heavy workflow is therefore a reasonable technical inference, but only as staffing evidence rather than confirmed product behavior (inferred from job postings) [Ground Floor Careers, August 2026] [Ashbyhq, November 2026].

What is not yet in public is as important as what is. The company has not publicly disclosed vessel-level output, commercial pricing, deployment site, named customers, or a verified demonstration of continuous power generation at sea in the materials reviewed [MLQ, November 2026] [Proofstory, November 2026]. For investors, that leaves the product story promising in concept but still dependent on future proof points in marine engineering, power conversion, uptime, and data-center operations under offshore conditions [TechCrunch, November 2026] [The Company Wire, November 2026].

One source, partially checked -- Product concept is corroborated across company materials and multiple press reports, but key technical performance claims remain pre-deployment and largely company-sourced.

The Market They Are Entering

Market context

PUBLIC The market matters now because Drogue is trying to sell into a real and visible constraint, computing demand is rising while power and cooling capacity for new data-center builds remains scarce in many markets, according to public reporting on the company and its category [TechCrunch, November 2026] [The Company Wire, November 2026].

The public record does not provide a third-party TAM, SAM, or SOM estimate specific to floating offshore data centers powered by ocean thermal energy conversion, and the available coverage stays at the concept level rather than sizing the addressable market [TechCrunch, November 2026] [MLQ, November 2026]. That matters because Drogue sits at the intersection of at least three larger markets, data-center capacity, power generation, and industrial cooling, but none of the cited sources quantify the overlap in a way that would support a clean market model from public evidence alone [TechCrunch, November 2026] [Proofstory, November 2026].

A narrower way to read the opportunity, using only the cited reporting, is as an attempt to serve data-center developers, operators, and other energy-intensive computing users that cannot secure land, interconnection, or cooling quickly enough through conventional onshore infrastructure [AI Market Watch, November 2026] [TechCrunch, November 2026]. That is a plausible demand wedge, but it remains a wedge, not a measured market segment in the public materials.

Market slice	Public evidence	Relevance to Drogue
Floating offshore data centers using OTEC	No third-party sizing identified in the cited public sources [TechCrunch, November 2026] [Proofstory, November 2026]	Direct category, but still pre-defined in public markets research
Data-center developers and operators facing power constraints	Reporting describes long interconnection and construction timelines affecting land-based AI infrastructure [TechCrunch, November 2026] [The Company Wire, November 2026]	Most credible near-term buyer set described in public reporting
Energy-intensive computing workloads, including AI	Drogue's public positioning centers on energy-intensive computing and AI workloads [TechCrunch, November 2026] [MLQ, November 2026]	Core workload category if the platform proves technically and commercially viable

The table is sparse because the evidence is sparse. The useful signal is not a neat market-size figure, but that the company is oriented toward a bottlenecked buyer problem that multiple public reports describe consistently, even if none quantify the spend pool.

Demand drivers and adjacent markets

PUBLIC The strongest tailwind in the public record is simple, AI-linked compute demand is colliding with the practical limits of power access and thermal management for onshore facilities [TechCrunch, November 2026] [The Company Wire, November 2026].

That framing shows up repeatedly in the launch coverage. TechCrunch describes Drogue's pitch as colocating generation and computing offshore, potentially sidestepping some grid constraints while using seawater as a heat sink [TechCrunch, November 2026]. The Company Wire similarly presents the company as a response to land-based AI infrastructure bottlenecks [The Company Wire, November 2026]. Taken together, the demand driver is less "offshore data centers" as a stand-alone category and more the scramble for new compute capacity where the grid is the gating factor.

The adjacent markets are easier to identify than the core market size. Drogue touches conventional data-center development, distributed power generation, maritime industrial systems, and large-scale cooling infrastructure, based on its stated plan to install a multi-megawatt prototype on a conventional ocean-going ship [MLQ, November 2026] [TechCrunch, November 2026]. Those adjacencies cut both ways, they enlarge the conceptual market, but they also import execution dependencies from sectors with long build cycles and heavy permitting.

Substitutes are also implicit in the reporting. If grid access improves or developers can secure faster onshore power and cooling, the value of a floating offshore architecture narrows to cases where offshore deployment offers a cost, speed, or resilience advantage, an inference grounded in Drogue's own stated wedge around interconnection and cooling rather than any disclosed customer contract [TechCrunch, November 2026] [Proofstory, November 2026].

Regulatory and macro forces

PUBLIC The macro case is tied to infrastructure scarcity, but the regulatory case is still underdeveloped in the public file.

The public reporting does not identify named deployment partners, a shipbuilding partner, a utility relationship, or a disclosed deployment site for the planned early-2028 prototype [TechCrunch, November 2026] [Proofstory, November 2026]. Without those details, it is difficult to map the exact permitting path, maritime compliance burden, or jurisdictional constraints from public evidence alone. That does not negate the opportunity, but it does mean the market is being pursued ahead of a clearly documented regulatory playbook in the public domain.

There is also a capital-markets angle. Drogue raised $20 million in November 2026 from a syndicate that includes Toyota Ventures, Boost VC, Slow Ventures, and XYZ Venture Capital, among others, which suggests investor appetite for infrastructure-compute convergence remains open at the seed stage even before commercial proof points are visible [TechCrunch, November 2026] [PIQ Markets, November 2026]. Still, the absence of disclosed customers or commercial partners means macro enthusiasm for AI infrastructure has not yet translated, in public, into validated demand for this specific architecture [Proofstory, November 2026] [MLQ, November 2026].

One source, partially checked -- Market framing is supported by named public reporting, but no independent third-party market sizing or regulatory documentation was identified in the cited sources.

The Competitive Field

Market structure

MIXED Drogue is not competing head-on with a single named startup peer in the available public record so much as it is proposing a different stack for the same constraint: how to secure power and cooling for energy-intensive compute without waiting on land-based grid buildout [TechCrunch, November 2026].

That distinction matters, because the competitive set is better understood in layers than in logos. At the incumbent layer sit conventional land-based data-center developers and operators, which remain the default procurement path for AI and other high-density workloads even as interconnection queues and power availability have tightened in key markets, according to Drogue's launch coverage [TechCrunch, November 2026]. At the challenger layer sits Drogue itself, with a ship-based platform that combines ocean thermal energy conversion with seawater cooling and aims to colocate generation and compute offshore [TechCrunch, November 2026] [MLQ, November 2026]. The adjacent substitutes are not named companies in the source set, but rather alternative ways to solve the same buyer problem: securing compute capacity through new land-based builds, retrofits, or power-first infrastructure arrangements rather than through floating assets [TechCrunch, November 2026].

The absence of publicly named direct peers cuts two ways. On one hand, it suggests Drogue may be early in a niche where category boundaries are still forming, which can help with narrative ownership and investor attention after a $20 million seed financing in November 2026 [TechCrunch, November 2026] [PIQ Markets, November 2026]. On the other, it means the company has not yet been benchmarked in public reporting against a clear set of operating comparables on deployment cost, uptime, siting, customer adoption, or regulatory throughput. For now, the public market map is conceptual rather than crowded by disclosed rivals.

Drogue's clearest edge today is not proven commercial traction, which has not been publicly disclosed, but the way its founding story aligns engineering, infrastructure timing, and fundraising. Public reporting ties Sam Sipe to engineering leadership at Scale AI and prior U.S. Navy service, a combination that is at least directionally relevant to mission-critical systems and maritime operating environments [Linear] [MLQ, November 2026]. The company also assembled a broad investor syndicate that includes Toyota Ventures, Slow Ventures, Boost VC, and XYZ Venture Capital, which may help with recruiting and follow-on financing as the prototype effort advances [TechCrunch, November 2026]. That edge is real but perishable: if Drogue does not convert early attention into a credible shipboard deployment path by its stated early 2028 target, the current advantage remains mostly narrative and capital access rather than a durable moat [TechCrunch, November 2026] [Runtimewire].

The exposure is easier to define than the moat. Drogue has not publicly disclosed a customer, offtake agreement, deployment site, shipbuilding partner, or utility-style counterparties, leaving it exposed to any conventional provider that can offer earlier certainty on timelines and service levels [Proofstory, November 2026] [TechCrunch, November 2026]. It is also exposed to categories it does not appear to own today, particularly established land-based deployment channels and customer procurement relationships. In practical terms, the company may be competing less against a named startup and more against the inertia and bankability of existing data-center procurement, where buyers can choose familiar infrastructure even if it is slower to energize.

Over the next 18 months, the most plausible competitive scenario is a split between concept leadership and execution leadership. Drogue is the most visible winner if offshore power-plus-cooling moves from idea to demonstrable prototype on the timetable management has discussed publicly, because no other named company in the source set currently owns that narrative position [TechCrunch, November 2026] [MLQ, November 2026]. Drogue is the most obvious loser if the early-2028 prototype slips materially or remains short on disclosed counterparties, because the market would then default back toward incumbent land-based options that, while not individually named here, already control the buyer relationship and operating trust. In other words, the near-term contest is not yet startup versus startup. It is novel architecture versus the installed base.

One source, partially checked -- This section relies primarily on TechCrunch and MLQ for Drogue's positioning and timeline, with Proofstory used to corroborate the absence of disclosed customers and partners.

Opportunity

Upside case

PUBLIC The prize here is unusually large because the company is not trying to sell a narrower data-center component, it is trying to collapse power generation, cooling, and compute siting into a single offshore asset for a category, AI infrastructure, that is now constrained as much by electricity and heat rejection as by chips [TechCrunch, November 2026] [MLQ, November 2026].

The headline opportunity is straightforward. If Drogue can put a multi-megawatt prototype on a conventional ocean-going ship in early 2028 and show that ocean thermal energy conversion can power and cool energy-intensive computing at sea, it could become an early platform provider for a new class of off-grid compute infrastructure rather than a niche hardware vendor [TechCrunch, November 2026] [MLQ, November 2026]. That outcome is still speculative, but it is not untethered from evidence: the company has already raised $20 million from a syndicate that includes Toyota Ventures, XYZ Venture Capital, Boost VC, and Slow Ventures, and public reporting describes a designed multi-megawatt prototype rather than a concept-only research effort [TechCrunch, November 2026] [MLQ, November 2026] [PIQ Markets, November 2026].

The evidence also suggests a wedge that is legible to customers. Public coverage consistently frames Drogue around a real bottleneck, land-based AI infrastructure faces long interconnection and construction timelines, and the company is positioning an offshore deployment model that colocates generation and cooling with compute demand [TechCrunch, November 2026] [The Company Wire, November 2026]. When a startup is addressing a timing constraint as much as a cost constraint, a first working deployment can matter more than a polished product catalog.

Scenario	What happens	Catalyst	Why it's plausible
First offshore AI capacity provider	Drogue proves its early-2028 shipboard prototype and becomes an initial supplier of floating, off-grid compute capacity for operators that cannot wait for terrestrial power interconnection	Successful deployment of the first multi-megawatt vessel prototype in early 2028	The company is specifically targeting a multi-megawatt shipboard deployment timeline, and public reporting ties the product to grid-constrained AI workloads [TechCrunch, November 2026] [MLQ, November 2026]
Platform for data-center developers and operators	Drogue sells or partners its ship-based power-and-cooling system into a broader class of data-center developers and operators rather than operating every vessel itself	A commercial partnership with a data-center operator or developer after prototype validation	Available reporting identifies the customer category as data-center developers and operators, and no public partnership has been named yet, leaving room for a platform model if the technical proof point lands [AI Market Watch, November 2026] [Proofstory, November 2026]
Offshore energy-compute stack	Drogue extends from one prototype into a repeatable vessel design and becomes an infrastructure stack that combines generation, cooling, controls, and onboard computing for remote or high-intensity workloads	Standardization of the vessel architecture after the first deployment and follow-on financing to scale manufacturing and deployment	The company is hiring across thermodynamics, electrical controls, and simulation, which is consistent with building an integrated system rather than a single subsystem [Ashbyhq, November 2026] [Ground Floor Careers, August 2026]

What compounding could look like is more industrial than software-native, but the logic is still there. A first working vessel would not only validate the thermodynamics, it would also generate operational know-how in ship integration, controls, thermal management, and deployment sequencing that later entrants would need to reproduce [TechCrunch, November 2026] [Ashbyhq, November 2026]. In deep infrastructure, process learning can be as defensible as code.

There is an early signal of that systems orientation in the hiring footprint. Drogue's public roles span thermodynamics, electrical and controls engineering, generalist engineering, and simulation, which points to a company trying to solve an interdependent stack where each iteration should improve design assumptions, vessel performance, and deployment confidence [Ashbyhq, November 2026] [Ground Floor Careers, August 2026]. If that loop starts working, each ship could make the next ship cheaper to design, easier to finance, and faster to deploy. That is the closest thing this model has to a flywheel, and it would matter more than brand at this stage.

The size of the win cannot be anchored to a confirmed market-size figure from the current source set, so the right way to frame it is by infrastructure role rather than by a fabricated TAM. If Drogue becomes the early category reference for offshore, self-powered AI data centers, the outcome could resemble a strategic infrastructure asset rather than a conventional hosting startup, especially because the system combines energy and compute in one platform [TechCrunch, November 2026]. A reasonable upside framing is that a successful first vessel, followed by repeatable deployments and operator partnerships, could support a venture-scale outcome in the billions of dollars in enterprise value (scenario, not a forecast). That inference rests on the company solving a binding constraint in AI infrastructure, not on any disclosed revenue today, and it remains contingent on technical proof that has not yet been demonstrated publicly [TechCrunch, November 2026] [Proofstory, November 2026].

One source, partially checked -- Section relies primarily on TechCrunch and MLQ, with partial corroboration from PIQ Markets, The Company Wire, Ashbyhq, Ground Floor Careers, AI Market Watch, and Proofstory. Core upside thesis remains pre-deployment and partly company-disclosed.

Sources