NanoForge AI

AI and robotics platform automating the full materials R&D workflow for advanced materials discovery.

Cover Block

From the public record

Field	Value
Name	NanoForge AI
Tagline	AI and robotics platform automating the full materials R&D workflow for advanced materials discovery. [nanoforgeai.com]
Headquarters	Seoul, South Korea [nanoforgeai.com]
Founded	2025 [Maeil Business Newspaper, July 2025]
Stage	Seed [The Asia Business Daily, July 2025]
Business Model	Hardware + Software
Industry	Deeptech
Technology	AI / Machine Learning
Geography	East Asia
Growth Profile	Venture Scale
Founding Team	Co-Founders (2): Kim Dong-hyun and Bae Jae-won [The Asia Business Daily, July 2025]
Funding Label	Seed
Total Disclosed Funding	$10.9 million [Dealroom.co, October 2026]

The Short Version

PUBLIC NanoForge AI is building an autonomous materials R&D platform that combines AI models with robotics to automate design, synthesis, and analysis, and it merits investor attention because it has moved from formation to seed financing and early industrial collaborations within roughly 18 months of founding [Maeil Business Newspaper, July 2025] [The Asia Business Daily, July 2025] [Dealroom.co, October 2026]. Public reporting places the company’s formation in April 2025 in Seoul, with co-founders Kim Dong-hyun and Bae Jae-won positioning the business around a closed-loop laboratory workflow rather than a software-only materials screening tool [Maeil Business Newspaper, July 2025] [The Asia Business Daily, July 2025] [nanoforgeai.com].

The product claim is straightforward, if still mostly company-described: NanoForge says its system generates candidate materials, evaluates stability and synthesizability, designs synthesis recipes, defines process windows, and optimizes experimental conditions through self-driving laboratories [nanoforgeai.com]. That integrated AI-plus-automation stack is also the company’s reported point of differentiation for customers pursuing advanced materials programs in batteries, semiconductors, mobility, ceramics, superconductors, and defense-related applications [The Asia Business Daily, July 2025] [nanoforgeai.com].

The founding team’s public record is early but directionally relevant. Kim is identified as co-founder and CEO, Bae as co-founder and CTO, and both were reported as graduates of the Korea Science Academy of KAIST, while external profile data links Kim to prior executive experience at HB SOLUTION and Bae to mathematics studies at KAIST [The Asia Business Daily, July 2025] [Bloomberg Markets] [Lacuna].

On financing, the company first surfaced with an undisclosed seed round from FuturePlay and Mashup Ventures in July 2025, then was reported to have raised KRW 14.6 billion, roughly $10.9 million, in October 2026 with Mirae Asset Venture Investment named as lead investor [The Asia Business Daily, July 2025] [Dealroom.co, October 2026] [The Asia Business Daily, October 2026]. The business model is best understood as hardware plus software, with value creation tied to automating experimental throughput for industrial and research organizations rather than selling a pure research SaaS product [The Asia Business Daily, July 2025] [nanoforgeai.com].

Over the next 12 to 18 months, the central watchpoints are whether reported partnerships convert into repeatable commercial programs, whether the EcoPro proof of concept and DNK Mobility relationship produce externally validated outcomes, and whether NanoForge can show that its closed-loop platform improves development speed or economics in real customer settings [MoneyToday, February 2026] [nanoforgeai.com]. The setup is promising, but most operating claims still rest on company materials or limited local press coverage, so the near-term investment case depends less on category excitement than on independent proof of adoption and technical execution [The Asia Business Daily, July 2025] [nanoforgeai.com].

Unconfirmed -- The section mixes independently reported funding and founder-role details with product and collaboration claims that remain primarily company-reported or lightly corroborated by local press.

Taxonomy Snapshot

Axis	Value
Stage	Seed
Business Model	Hardware + Software
Industry / Vertical	Deeptech
Technology Type	AI / Machine Learning
Geography	East Asia
Growth Profile	Venture Scale
Founding Team	Co-Founders (2)
Funding	Seed, total disclosed approximately $10.9 million

The Company in Brief

PUBLIC NanoForge AI appears to have formed in Seoul in 2025, with public materials describing an autonomous materials R&D company that combines AI models with laboratory automation to shorten the path from material design to experimental validation [Crunchbase] [nanoforgeai.com]. Maeil Business reported the company was established in April 2025, but this section stays with the website and Crunchbase record for the core corporate outline because the sourcing standard here is narrower [Crunchbase] [nanoforgeai.com].

The public footprint is still early, which is consistent with a newly formed seed-stage deeptech company. Crunchbase lists NanoForge AI as a Seoul, South Korea based company, while the company website frames the business around a full materials discovery workflow rather than a single point tool, positioning the platform across design, synthesis, and analysis inside a closed-loop lab system [Crunchbase] [nanoforgeai.com].

On milestones, the public record confirms a 2025 founding and a later disclosed seed financing of roughly $10.9 million, or KRW 14.6 billion, by October 2026, which marks the company’s first clearly visible capital formation event in structured public databases [Crunchbase]. The website also indicates the company had begun presenting customer-facing collaboration activity by mid-2026, although those operating details are better assessed alongside product and traction evidence than as corporate formation facts [nanoforgeai.com].

Single-source, plausible -- Confirmed primarily by Crunchbase and the company website; founding month and several milestone details require broader corroboration outside the sources permitted for this section.

What They Have Built

MIXED NanoForge AI is positioning around workflow coverage rather than a single model claim. On its website, the company describes a closed-loop "Design, Synthesis, and Analysis" system built around self-driving laboratories, with AI used to generate material candidates, evaluate stability and synthesizability, design synthesis recipes, define process windows, and optimize experimental conditions [nanoforgeai.com]. Press coverage broadly matches that framing, describing an autonomous digital materials lab that combines AI-based material design with robotics-led synthesis and feedback [The Asia Business Daily, July 2025] [Maeil Business Newspaper, July 2025].

The practical point is that NanoForge AI is selling an integrated hardware-plus-software stack into advanced materials R&D rather than a pure discovery model. The July 2025 reporting says the target buyer is organizations developing new materials for batteries, semiconductors, mobility, and other high-performance applications, and it attributes the company's differentiation to combining AI prediction with automated synthesis and process optimization [The Asia Business Daily, July 2025]. The company website lists initial application areas that include superconductors, batteries, semiconductors, ceramics, and defense materials, but those domain claims remain company-reported rather than independently validated [nanoforgeai.com].

A few operating details are visible, but the external evidence still stops short of a verified technical demo or published performance benchmark. NanoForge AI's careers page organizes work across Design, Process, and AutoLab functions, which is directionally consistent with the product architecture described on the website [nanoforgeai.com]. Public sources do not establish model architecture, laboratory throughput, synthesis success rates, or whether any customer deployment is running in routine production conditions, so the technical case today rests more on product concept coherence than on disclosed benchmark data [MoneyToday, February 2026] [nanoforgeai.com].

Unconfirmed -- Core product capabilities rely primarily on company website claims, with partial framing corroborated by The Asia Business Daily and Maeil Business Newspaper.

Market Size and Demand

PUBLIC The market matters now because the cost and strategic value of advanced materials development are both rising, and NanoForge AI is positioning itself inside that bottleneck rather than at the edge of it [The Asia Business Daily, July 2025] [Maeil Business Newspaper, July 2025].

Public source coverage does not provide a direct TAM, SAM, or SOM for NanoForge AI’s category, so the cleaner way to frame the addressable market is through adjacent spend pools that the company explicitly targets: batteries, semiconductors, mobility, ceramics, superconductors, and defense-related materials research [nanoforgeai.com] [The Asia Business Daily, July 2025]. That matters because the company is not selling a single-point software tool for simulation alone, according to its website and local press coverage, but a workflow that combines AI-led materials design with automated synthesis and analysis in a closed loop [nanoforgeai.com] [Maeil Business Newspaper, July 2025]. In practical terms, the relevant budget lines likely sit across industrial R&D, lab automation, and materials informatics, though the available public evidence does not let this report quantify those segments with precision.

The demand signal in the reporting is narrower than a broad market-sizing exercise, but it is directionally consistent. The Asia Business Daily described NanoForge AI as selling to organizations pursuing advanced materials research for batteries, semiconductors, mobility, and other high-performance applications, with differentiation tied to combining AI prediction with automated synthesis and process optimization [The Asia Business Daily, July 2025]. Maeil Business Newspaper separately reported the company’s ambition to shorten development cycles from years to weeks, which is a familiar procurement logic in materials R&D where iteration speed can matter as much as model accuracy, especially for corporate labs trying to move from candidate discovery to manufacturable process conditions [Maeil Business Newspaper, July 2025].

Adjacent and substitute markets are also important here because they define both upside and competition for budget. On one side are materials simulation software, informatics platforms, electronic lab systems, and contract research workflows that address only parts of the discovery stack. On the other are internal R&D teams at conglomerates and industrial labs that may prefer to build bespoke automation infrastructure rather than buy a third-party platform. NanoForge AI’s own product description suggests it is trying to collapse those layers into one operating system for materials experimentation, which broadens theoretical scope but also means the company is asking buyers to change process, not just vendor, behavior [nanoforgeai.com].

Regulatory and macro forces in the public record are implicit rather than quantified, but they still point in a supportive direction. The company’s cited application areas, batteries, semiconductors, mobility, and defense materials, align with sectors where supply-chain resilience, domestic manufacturing capability, and strategic technology independence have become more salient for governments and large industrial groups in East Asia and beyond [nanoforgeai.com] [The Asia Business Daily, July 2025]. Its recognition in KAIST’s E*5 program and the K-DeepTech Startup King of Kings competition also suggests a policy environment in South Korea that is receptive to deep-tech commercialization, even if awards are not the same thing as end-market adoption [The Asia Business Daily, July 2025] [MoneyToday, February 2026].

Market lens	What public sources support	Relevance to NanoForge AI
Advanced materials R&D workflows	Buyers include organizations developing materials for batteries, semiconductors, mobility, and other high-performance applications [The Asia Business Daily, July 2025]	Defines the practical demand environment more clearly than a single category label
Lab automation and self-driving labs	The platform is described as a closed-loop design, synthesis, and analysis system powered by self-driving laboratories [nanoforgeai.com]	Suggests exposure to automation budgets, not only software budgets
Materials informatics and AI-led discovery	The company says its system generates candidates, evaluates stability and synthesizability, and optimizes experimental conditions [nanoforgeai.com]	Positions the product within AI-enabled discovery tooling, but with broader scope
Strategic industrial R&D	Publicly cited application areas include superconductors, batteries, semiconductors, ceramics, and defense materials [nanoforgeai.com]	Expands the possible buyer set, while also increasing complexity of sales and validation

The table shows why this market should be read as a convergence story rather than a clean software category. The opportunity is probably larger than a pure materials-informatics niche, but the burden of proof is higher because the company has to win against existing lab processes and internal tooling, not just direct vendors.

Unconfirmed -- Market framing relies mainly on company website claims and one local news report; no independent third-party TAM, SAM, or SOM figures were identified in the provided public sources.

Who Else Is Fighting for This

MIXED NanoForge AI appears to sit in the narrow band between software-first materials informatics and capital-intensive lab automation, with its public pitch centered on joining AI-driven design to robotic synthesis in one workflow rather than selling either layer alone [nanoforgeai.com] [The Asia Business Daily, July 2025].

The competitive map is easier to define by substitute category than by a clean set of named startup peers, because the available source set does not identify direct competitors by name [The Asia Business Daily, July 2025]. On one side are incumbent internal R&D teams at industrial materials companies, which already run discovery programs for batteries, semiconductors, ceramics, and related high-performance materials, often with established process know-how and procurement budgets even if their workflows remain slower and more manual [The Asia Business Daily, July 2025]. On another side are software-led materials discovery tools, the kind of platform that can propose candidates or model properties but stops short of owning synthesis and feedback loops; NanoForge AI is explicitly trying to differentiate from that category by integrating prediction, automated synthesis, and process optimization [The Asia Business Daily, July 2025] [nanoforgeai.com]. The third substitute is outsourced experimentation through research institutes, university labs, and contract development relationships, especially in Korea's deeptech ecosystem, where technical credibility and founder networks can matter as much as product breadth [The Asia Business Daily, July 2025] [MoneyToday, February 2026].

The company's defensible edge, at least on public evidence, is not brand or distribution yet. It is the attempt to control the full experimental loop, which the company describes as a closed-loop "Design, Synthesis, and Analysis" system powered by self-driving laboratories [nanoforgeai.com]. If that architecture works in production settings, the advantage could compound through proprietary experimental data, faster iteration, and tighter tuning of synthesis recipes and process windows over time [nanoforgeai.com]. The problem is that this edge still looks perishable until there is external proof of repeat commercial use: the DNK Mobility relationship and EcoPro proof of concept suggest early industrial interest, but neither source establishes scaled deployment, recurring revenue, or switching costs today [MoneyToday, February 2026] [nanoforgeai.com].

That leaves NanoForge AI most exposed to organizations that already own either customer access or physical lab infrastructure. FuturePlay, Mashup Ventures, and Mirae Asset Venture Investment provide financing credibility, but capital raised does not by itself solve go-to-market, validation, or hardware utilization risk [The Asia Business Daily, July 2025] [Dealroom.co, October 2026]. Internal R&D groups at large materials and manufacturing companies can remain the default option if they decide that externalizing sensitive discovery work is unnecessary, especially in defense, semiconductors, or strategic battery materials where process IP may be guarded closely [nanoforgeai.com] [The Asia Business Daily, July 2025]. The adjacent software-only category is also a pressure point: if customers value candidate generation and simulation more than automated wet-lab execution, a lighter-weight tool could be easier to adopt than a fuller-stack platform that requires trust in both algorithms and robotics.

The most plausible 18-month scenario is a sorting between platforms that can show repeatable industrial proof points and those that remain technically interesting but procurement-light. On the public record, EcoPro is the named party most likely to matter as a "winner if X" case: if the reported battery-material proof of concept converts into a broader commercial program, NanoForge AI would have a reference account in one of its priority verticals and a stronger claim that the integrated workflow matters beyond demos [nanoforgeai.com]. The more vulnerable side of the ledger is the DNK Mobility line of business if aluminum heat-dissipation work stays confined to partnership language without visible follow-on scope; in that case, the likely loser is not necessarily NanoForge AI as a company, but its argument that early collaborations are turning into defensible market position [MoneyToday, February 2026].

Unconfirmed -- This section relies primarily on company claims and a small number of named-publisher reports; the available public sources do not identify direct competitors by name or independently validate commercial deployment depth.

Opportunity

PUBLIC The prize here is not a better lab software tool, it is the chance to become a system of record and execution layer for industrial materials discovery in categories where a shorter R&D cycle can matter to batteries, semiconductors, thermal management, and defense supply chains [The Asia Business Daily, July 2025] [nanoforgeai.com].

The headline opportunity is straightforward: NanoForge AI could become a category-defining platform for closed-loop materials R&D if it can convert its current positioning from founder-led technical credibility into repeatable enterprise adoption. That outcome is at least reachable on the public record because the company is not claiming a narrow prediction engine. It presents a combined stack of AI-based candidate generation, synthesis planning, process-window design, and robotic experimentation, which is closer to workflow replacement than point-solution augmentation [nanoforgeai.com]. The early evidence is still thin, but it is directionally relevant: seed backing first surfaced from FuturePlay and Mashup Ventures in July 2025, then a larger seed round of KRW 14.6 billion, roughly $10.9 million, was reported in October 2026 with Mirae Asset Venture Investment identified as lead [The Asia Business Daily, July 2025] [Dealroom.co, October 2026]. Publicly reported collaboration signals with DNK Mobility and EcoPro also suggest the company is being pulled toward real materials programs rather than remaining a purely academic platform [MoneyToday, February 2026] [nanoforgeai.com].

The upside path is still scenario-driven, not linear. The table below frames the most plausible routes to outsize value from the evidence currently available.

Scenario	What happens	Catalyst	Why it's plausible
Industrial co-development platform	NanoForge AI becomes the preferred external platform for manufacturers that need faster iteration on specialty materials programs, starting with battery and thermal-management use cases	A reported collaboration converts into a broader multi-program relationship, especially if EcoPro or DNK Mobility expands scope beyond a proof of concept or single initiative [nanoforgeai.com] [MoneyToday, February 2026]	The company is already presenting sector-specific applicability in batteries, semiconductors, mobility, ceramics, and defense, and at least two named organizations have been linked publicly to technical work [The Asia Business Daily, July 2025] [MoneyToday, February 2026]
Autonomous lab operating system	The company standardizes the "Design, Synthesis, and Analysis" loop into software-plus-automation infrastructure used by corporate and institutional labs across multiple materials domains	A reference deployment demonstrates that the integrated AI-plus-robotics workflow shortens cycle time enough to justify platform adoption over fragmented incumbent tools [nanoforgeai.com] [Maeil Business Newspaper, July 2025]	Its stated differentiation is the integration of prediction, automated synthesis, and process optimization, which is the kind of broader control point that can support infrastructure status if technical results hold up [The Asia Business Daily, July 2025] [nanoforgeai.com]
Korea-to-global deeptech supplier	NanoForge AI uses South Korean industrial relationships and local recognition to become an exportable deeptech vendor for advanced manufacturing R&D teams abroad	A high-credibility domestic customer outcome, backed by continued investor support after the 2026 seed round, becomes the reference case for expansion [Dealroom.co, October 2026] [MoneyToday, February 2026]	The company has already won visible domestic startup recognition, including second place in KAIST's E*5 program and a top award in the 2025 K-DeepTech Startup King of Kings competition, which can help with talent, partnerships, and enterprise access in its home market [The Asia Business Daily, July 2025] [MoneyToday, February 2026]

The compounding logic, if it works, comes from learning loops rather than classical consumer network effects. A platform that proposes candidates, runs synthesis, and feeds results back into the next cycle should improve as it accumulates more experiment-output pairs across adjacent materials problems [nanoforgeai.com]. If NanoForge AI can keep those loops inside customer workflows, one successful battery or thermal-materials program could do more than generate revenue. It could deepen proprietary process knowledge, sharpen recipe design, and make subsequent projects faster to start and harder to displace. The early signs of that flywheel are modest but visible in the company's own product framing and in the fact that reported use cases already span heat-dissipation materials and battery materials rather than a single narrow domain [MoneyToday, February 2026] [nanoforgeai.com].

The size of the win is best framed qualitatively because the source set here does not include a confirmed market-sizing study or a named public comparable with directly cited valuation data. Even so, the contours are clear enough to state the upside in concrete terms. If the autonomous lab operating system scenario plays out, and NanoForge AI becomes a trusted external R&D platform across a handful of major industrial categories, the company could plausibly grow into a strategic infrastructure asset for advanced manufacturing research rather than a project-based tools vendor (scenario, not a forecast). The public evidence does not support a hard valuation number yet, but the combination of a full-stack technical claim, early enterprise-adjacent collaboration signals, and nearly $10.9 million in disclosed seed financing is enough to treat the upside as institutionally relevant rather than merely conceptual [Dealroom.co, October 2026] [The Asia Business Daily, July 2025] [MoneyToday, February 2026].

Unconfirmed -- This section relies on a mix of independent funding and news coverage, but key product capabilities and one collaboration signal are company-reported and not yet independently validated.

Sources