MakinaRocks
MakinaRocks is a Korean publicly-listed industrial AI software company that builds the Runway AI operating system and Forward Deployed Engineering services for deploying AI in closed-network manufacturing, defense, semiconductor, battery and public-infrastructure environments, serving major Korean conglomerates and government clients.
- Company typePublic
- Founded2017
- HeadquartersSeoul, South Korea
- Headcount101–250
- GTM typeB2B
- OfferingSoftware
What MakinaRocks does
MakinaRocks is a Korean industrial AI software company founded in 2017 that builds and operates the Runway AI operating system — a Kubernetes-based MLOps/LLMOps platform designed to deploy, orchestrate and operate AI in closed-network industrial environments. Headquartered in Seoul with a Tokyo office opened in April 2025, the company serves Korean conglomerates (Samsung, Hyundai, SK, LG, POSCO, Hanwha, Doosan) and public-sector clients (K-water, Korea Insurance Development Institute) in mission-critical verticals including automotive, semiconductors, batteries, electronics, heavy industry, defense and public infrastructure. Its product portfolio includes the Runway AI OS core platform, the DrawX drawing-AI agent solution, Runway Annotation Studio for industrial data labeling, and Runway App Studio for vibe-coded AI agent prototyping, supplemented by a portfolio of vertical AI agents (expert augmentation, control, predictive operations, machine vision) and planned Dark Factory OS and Defense OS products funded by IPO proceeds.
The company operates a Forward Deployed Engineering (FDE) model with approximately 50 AI engineers, project managers and software engineers embedded at customer sites to co-build production AI systems, executing 123 projects in 2025 alone. Its proprietary technology stack includes Compound AI Systems, a Multi-Step AI Digital Twin, a VLM-based RFQ automation engine, and multi-agent Physical AI for autonomous PLC control in production lines. The company generates revenue from a mix of Runway platform subscriptions (on-premise license and cloud SaaS) and FDE professional services, with quote-based multi-year contracts and a PoC-led sales motion. As of May 20, 2026 MakinaRocks is publicly listed on KOSDAQ (477850), with cumulative funding of approximately $60M+ from strategic investors including Hyundai (ZER01NE), Samsung, LG, Hanwha, POSCO, Applied Ventures, KDB and Positive Investment, and a stated revenue target of KRW 100 billion by 2030.
MakinaRocks firmographics
Firmographics- Name
- MakinaRocks
- Legal name
- ㈜마키나락스 (MakinaRocks Inc.)
- Website
- https://makinarocks.ai
- Company type
- Public
- Founded year
- 2017
- Operating status
- Ipo
- Headcount range
- 101–250 employees
- Short description
- MakinaRocks is a Korean publicly-listed industrial AI software company that builds the Runway AI operating system and Forward Deployed Engineering services for deploying AI in closed-network manufacturing, defense, semiconductor, battery and public-infrastructure environments, serving major Korean conglomerates and government clients.
- Ownership category
- akta.pro rank
MakinaRocks industry classification
Industry- Product category
- Industrial AI Software (MLOps/LLMOps platform and vertical AI agents for manufacturing)
- akta.pro primary industry
- End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management) (HDAEANAA)
- akta.pro secondary industries
- Model Deployment, Serving & Inference Platforms (HDAAABAF), Agents & Autonomous Workflows (Tool Use, Planning, Multi-Agent) (HDAAACAF), AI Observability, Monitoring & Evaluation Platforms (Drift, Quality, Safety) (HDAEANAF), AI/ML Engineering & Model Development Services (BPAEAHAF), Model Serving, Inference & Deployment Platforms (APIs, Edge/On-Prem) (HDAEANAC), AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs) (HDAEANAJ)
Keywords
Where MakinaRocks is headquartered
LocationHeadquarters
- HQ city
- Seoul
- HQ country
- South Korea
- HQ region
- Asia
Offices2 records
Markets served
MakinaRocks business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Personnel, Technology or R&D, Operations, Marketing or Sales, Infrastructure
Revenue model
- Runway (MLOps/LLMOps Platform): Enterprise MLOps/LLMOps platform offered as software product. Customers deploy on-premise (especially for closed-network environments) or cloud; sold with 'professional' and 'Cloud SaaS' expansion. References to 'selling' and contracts with enterprise clients.
- FDE Project / ML Solutions Delivery: Project-based revenue from Forward Deployed Engineering engagements — 123 projects executed in 2025 across semiconductors, energy, batteries and defense. Includes PoC fees, deployment services and ongoing operational support for AI applications.
- DrawX (Drawing AI Solution): Drawing AI agent solution that automates search, comparison, analysis and judgment of engineering drawings — sold to manufacturing enterprises.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Other | Multi-year contract | Quote-based / not publicly disclosed |
Go-to-market motion3 records
Distribution channels5 records
Marketing channels10 records
MakinaRocks product offering
Product offeringCore offering
MakinaRocks fundamentally develops and sells the Runway enterprise AI operating system — a Kubernetes-based MLOps/LLMOps platform for on-premise and cloud environments — together with the DrawX drawing-AI agent, the Runway Annotation Studio labeling tool, and a portfolio of specialized industrial AI agents (Expert Augmentation, Control, Predictive Operations, Machine Vision). The company delivers these products through a Forward Deployed Engineering model that embeds AI engineers, project managers and software engineers at customer sites to co-build production AI in closed-network environments across automotive, semiconductors, batteries, defense, heavy industry and public infrastructure.
Product overview
MakinaRocks offers a platform-plus-modules architecture centered on its AI operating system 'Runway,' which serves as the core enterprise AI platform for industrial AI development, deployment, and operations. Runway operates as a Kubernetes-based MLOps platform supporting the full ML lifecycle, with built-in features for experiment management, training, deployment, and serving, available for both on-premise and cloud environments. Built on top of Runway, MakinaRocks offers specialized add-on modules including 'DrawX' for engineering drawing AI automation, 'Runway App Studio' for vibe coding and AI agent prototyping, and 'Runway Annotation Studio' for industry-specific data labeling. The portfolio is rounded out by specialized AI Agents organized into four categories (Expert Augmentation, Control, Predictive Operations, and Machine Vision) targeting the automotive, defense, heavy industry, and manufacturing sectors. Planned future products include the 'Dark Factory OS' for fully automated manufacturing and 'Defense OS' for closed-network defense environments, both to be developed with IPO proceeds. Together, these products form a vertically integrated industrial AI stack designed to bring AI capabilities from development to production deployment in mission-critical industrial settings.
Differentiator
Problem solved
Functional benefit
Brands
- Runway: AI operating system (AI OS) for enterprises that rapidly scales, operates and orchestrates industrial-specialized AI across enterprise systems; described as a proprietary closed-network AI platform with 6,000+ deployed AI models and 25+ TB of operational data.
- DrawX
- Runway Annotation Studio
Products and services
- Runway (AI OS) Runway is MakinaRocks' enterprise AI operating system (AI OS) — a Kubernetes-based MLOps/LLMOps platform for industrial AI with a click-based UI for model deployment, operation, retraining, experiment management, training, model registration, deployment and serving, deployable on-premise (including closed-network / air-gapped environments) and on cloud. Target customers are large industrial enterprises, defense agencies, and public-sector organizations that need to operate AI inside closed networks, with 6,000+ AI models and 25+ TB of operational data already deployed on the platform.
- DrawX DrawX is a standalone AI agent solution for engineering drawing decision-making that converts drawing data into standardized corporate assets and automates search, comparison, analysis and judgment of drawings. It is built on VLM and multi-agent architecture and is sold to manufacturing enterprises for RFQ (Request-for-Quotation) automation, design document change review and drawing version management, handling 2D and 3D drawings.
- Runway Annotation Studio Runway Annotation Studio is industry-specific annotation / data-labeling software that integrates with Runway to deliver segmentation-based auto-annotation for manufacturing image datasets, unifying dataset creation and model operation workflows and replacing manual local-tool labeling.
- Forward Deployed Engineering (FDE) Services Forward Deployed Engineering (FDE) is a paid on-site delivery service in which ~50 AI engineers, project managers and software engineers are embedded at customer sites to co-build and complete AI production systems in semiconductors, energy, batteries and defense. 123 FDE projects were executed in 2025 and 20+ multi-agent AI solutions were commercialized under this motion, including PoC fees, deployment services and ongoing operational support.
- MRX Drawing Inspector MRX Drawing Inspector is an AI-based drawing interpretation, comparison and search solution for intelligent drawing operations, positioned as a vertical AI product for engineering drawing workflows.
- Specialized AI Agents (Expert Augmentation, Control, Predictive Operations, Machine Vision) Specialized AI Agents is a product family of industry-specialized vertical AI agents organized into four categories — Expert Augmentation (전문가 증강), Control (제어), Predictive Operations (예측 및 운영), and Machine Vision (머신비전) — designed for high-accuracy/reliability industrial use cases in automotive, defense, heavy industry and manufacturing. The agents are built on the company's compounded LLM and form the runtime layer of the Physical AI / Software Defined Factory stack.
Quantifiable outcome
- 50% RFQ lead time reduction (avg. 15 days → <7 days) and >90% reduction in manual drawing search time at a semiconductor equipment manufacturer
- +6 more outcomes
Companies that use MakinaRocks
Customer profileNamed customers20 records
Segments6 records
Ideal customer profiles4 records
MakinaRocks technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration9 records
AI capability16 records
Feature9 records
MakinaRocks partnerships and signals
Strategic signalPartnerships
Eleven partnerships are on record, tiered flagship and core.
- NVIDIAflagshipFeatured speaker at ATTENTION 2025 conference discussing Physical AI opportunities including Jetson Thor, COSMOS, and GR00T — indicating close technology alignment. GPU servers underpin MakinaRocks' on-premise AI platforms for customers such as K-water.
- LG / LG AI ResearchflagshipJoint participation at ATTENTION 2025 with LG AI Research co-research chief Um Woohyung presenting keynote on Agentic AI ecosystem evolution with LG Group internal use cases (schedule automation, cosmetic materials R&D, ETF portfolio optimization). LG Electronics Production Technology Institute's Park Seokwoo also presented AXplorer smart factory AI case.
- UpstagecoreCo-panelist at ATTENTION 2025 — Upstage CTO Lee Hwal-suk discussed AI applications and multimodal services alongside MakinaRocks CBO Heo Youngshin and others.
- FuriosaAIcoreCo-panelist at ATTENTION 2025 — FuriosaAI SVP Chung Youngbeom participated in panel on AI semiconductors and Agentic AI, marking strategic AI silicon partnership context.
- RealWorld (리얼월드)coreCo-panelist at ATTENTION 2025 — RealWorld CBO Lee Kangwook discussed robotics and embodied AI alongside MakinaRocks leadership.
- Korea Military Academy (육군사관학교)flagshipProfessor Kim Jonghwan presented 'AI-based firepower planning recommendation' at ATTENTION 2025, indicating collaborative research and joint defense AI development with MakinaRocks.
- Korea Ministry of Science and ICT (한국과학기술정보통신부)flagshipConvened industry leaders and foreign diplomats at World IT Show 2026 in Seoul to discuss 'physical AI' development. MakinaRocks participated in roundtable with Vice Minister Ryu identifying technology and deployment gaps; ministry pledged continued policy support, data infrastructure expansion and backing for large-scale pilot projects.
- Hyundai CRADLE (현대크래들)coreHyundai Motor Group's global open innovation organization (offices in Silicon Valley, Tel Aviv, Berlin, Singapore) co-hosts events such as MIF (Mobility Innovators Forum) in San Francisco where MakinaRocks participates; collaborates with ZER01NE Ventures on global startup sourcing and opportunities.
- HubSpot, Inc.coreCRM and newsletter platform provider — MakinaRocks transfers customer information (name, company, dept, title, phone, email) to HubSpot in the United States for CRM operations, marketing automation and newsletter delivery (Stevie Co. also used for newsletter distribution).
- K-water (Korea Water Resources Corporation) — Waterround consortiumflagshipStrategic partnership starting 2024 via K-water's open digital water platform 'Waterround' consortium agreement. Includes 'MLOps-based data analysis model operation system' project and AI/AX-specific data analytics education. Currently co-developing AI-OS 'Runway'-based generative AI assistant for smart water treatment plant operations across 44 plants nationwide.
- Spacebank — Defense AX Council (방산 AX 협의회)coreMakinaRocks participates in the Defense AX Council at Pangyo (led by Spacebank) bringing together defense firms, ICT companies, AI firms, universities and public agencies to accelerate AI transformation in the defense sector with planned ~KRW 30 billion funding.
Scale indicators11 records
Recent moves8 records
Expansion highlights6 records
MakinaRocks competitors and assessment
Company assessmentBroad incumbents
- Senseye (Siemens): Predictive maintenance AI for industrial machinery, now part of Siemens; broader incumbent overlap with MakinaRocks' predictive operations agent capability but operating within a much larger industrial software portfolio.
- Cognex: Established machine vision vendor for factory automation and quality inspection; broad incumbent whose industrial vision capability overlaps with MakinaRocks' machine-vision agents, though from a hardware-centric base.
Direct peers
- C3.ai: Enterprise AI application platform selling industry-specific AI apps to manufacturing, energy and defense customers; closely comparable in vertical AI focus, on-prem/hybrid deployment and large-enterprise go-to-market.
- Uptake: Industrial AI platform for asset performance and predictive maintenance originally serving rail, manufacturing and energy; comparable in FDE-style industrial AI delivery and on-prem deployment model.
- Augury: Industrial AI for machine health monitoring and process optimization in manufacturing; comparable sensor-driven predictive maintenance capability referenced by MakinaRocks in motor downtime and Hyundai RPMS use cases.
- Braincube: Manufacturing AI platform combining IIoT data, MLOps and process optimization for industrial clients; comparable on factory-floor AI lifecycle management and reusable industrial AI templates.
- Palantir Technologies: Enterprise AI OS for government, defense and industrial clients with on-prem Foundry deployments; the most cited analog for MakinaRocks (nicknamed 'Korean Palantir') given vertical specialization, air-gapped deployment and agentic AI capabilities.
- Landing AI: Manufacturing-focused computer vision and AI platform (LandingLens) for visual inspection in factory environments; directly comparable on industrial computer vision and AI-on-the-production-line use cases.
Emerging players
- Falkonry: Industrial AI for time-series anomaly detection and predictive operations in process manufacturing; comparable focused use case for time-series AI on factory sensor data, but at smaller scale than MakinaRocks.
Regional players
- Preferred Networks: Japan-based industrial AI and deep-learning company focused on manufacturing, robotics and autonomous systems; regional peer with overlapping industrial AI focus and shared Japanese enterprise customer base.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat6 records
Key risks6 records
Key highlights7 records
Customer concentration
MakinaRocks social profiles
Digital presenceMakinaRocks compliance and trust
Trust signalCompliance7 records
MakinaRocks financial estimates
Financial estimateRevenue estimate
Valuation estimate
MakinaRocks leadership team
Management profileNumber of profiles
Profiles9 records
MakinaRocks subsidiaries and ownership
Company hierarchySubsidiaries1 record
MakinaRocks funding detail
Funding detailFunding overview
Funding rounds6 records
Investors24 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
MakinaRocks M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about MakinaRocks
What does MakinaRocks do?
MakinaRocks fundamentally develops and sells the Runway enterprise AI operating system — a Kubernetes-based MLOps/LLMOps platform for on-premise and cloud environments — together with the DrawX drawing-AI agent, the Runway Annotation Studio labeling tool, and a portfolio of specialized industrial AI agents (Expert Augmentation, Control, Predictive Operations, Machine Vision). The company delivers these products through a Forward Deployed Engineering model that embeds AI engineers, project managers and software engineers at customer sites to co-build production AI in closed-network environments across automotive, semiconductors, batteries, defense, heavy industry and public infrastructure.
Is MakinaRocks a public or private company?
MakinaRocks is a public company. It is classified as public and is currently ipo.
When was MakinaRocks founded?
MakinaRocks was founded in 2017. It employs 101 to 250 people.
Where is MakinaRocks based?
MakinaRocks is headquartered in Seoul, South Korea, in the Asia region.
How does MakinaRocks make money?
Three revenue lines are on record. Runway (MLOps/LLMOps Platform) is the primary driver. The others are FDE Project / ML Solutions Delivery and drawX (Drawing AI Solution).
Who are MakinaRocks's main competitors?
Broad incumbents on record are Senseye (Siemens) and Cognex. Direct peers are C3.ai, Uptake, Augury, Braincube, Palantir Technologies and Landing AI. Falkonry is listed as an emerging player. Preferred Networks is listed as a regional player.
Does MakinaRocks have an API?
Yes. MakinaRocks' Runway platform provides API serving capabilities for deployed models, including gRPC (Google Remote Procedure Call) for high-throughput inference and traditional REST API serving. The customer interview with Korea Insurance Development Institute (KIDI) confirms that Runway serves approximately 20 AI models processing ~7 million daily transactions with 1.5-second response times via gRPC. The platform also supports Batch serving for processing large volumes of data at once. Authentication, rate limits, and versioning specifics are not publicly documented.
What industry is MakinaRocks in?
MakinaRocks's product category is Industrial AI Software (MLOps/LLMOps platform and vertical AI agents for manufacturing). Its primary akta.pro industry code is HDAEANAA, End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management), with a secondary code of HDAAABAF, Model Deployment, Serving & Inference Platforms.