Artificial intelligence (AI) is a technology that artificially implements human learning, reasoning, and perception capabilities—essentially enabling computers to act like a human brain. Traditional machine learning relies on skilled developers to design models that learn based on provided data. However, advancements in technology have made it possible for AI to autonomously create the optimal AI model (AutoML) and automatically manage the generated model (MLOps). The development of these two technologies is rapidly accelerating the sophistication of AI development.
[Fig] Status of AI Adoption and Application Fields
An AutoML solution is “AI that creates AI.” AutoML autonomously extracts data features, selects suitable algorithms, performs learning, and evaluates the resulting models to generate the AI model best suited for the target task. Specifically, it can automatically perform data design, preprocessing, feature identification, feature engineering, algorithm selection, hyperparameter optimization (HPO), model evaluation, and generate improved AI models without manual intervention.
[Fig] Overview of AutoML Technology
An MLOps solution is “AI that manages AI.” MLOps combines machine learning (ML) and operations (Ops) into one integrated system. It provides the management framework necessary to transition conventional research-stage ML models into production-ready ML models, simplifying processes and ensuring maintenance. MLOps automates model maintenance, retraining, deployment, and monitoring—enabling continuous improvement and operational stability of AI systems.
[Fig] Overview of MLOps Technology
Korean MLOps company DeepAuto Co., Ltd. (CEOs: Sungjoo Hwang and Wonyong Jung) successfully raised seed investment from Naver D2SF (D2 Startup Factory). DeepAuto’s MLOps solution automates the full AI lifecycle—data preprocessing, model search and optimization, and deployment—across enterprises adopting AI. In particular, its proprietary AutoMLOps framework has demonstrated superior efficiency compared to big tech solutions. Based on this technological capability, DeepAuto was recently selected for the Deep Tech TIPS program, securing 1.5 billion KRW in funding. The company has already secured partners such as Samsung Advanced Institute of Technology, Cheil Worldwide, and StradVision, and is currently recruiting for its U.S. expansion.
[Fig] DeepAuto Co., Ltd.’s Partner Network
Other Korean companies are also actively developing AutoML and MLOps-related technologies. TEN Co., Ltd. (CEO: Sejin Oh) develops customized MLOps solutions that utilize AI learning data and assetization technology to automatically generate and trade customized learning metadata. The company raised 3 billion KRW in a Pre-A funding round in recognition of this technology. Meanwhile, DSLab Global Co., Ltd. (CEO: Seungki Yeo) has developed auto-labeling technology, which combines automated data generation and labeling techniques to enable even non-experts to develop AI models easily. As such, the development of AutoML and MLOps technologies—AI that creates and manages AI—is progressing actively in Korea.
[Fig] TEN Co., Ltd.’s MLOps Solution Model and DSLab Global’s AutoML Structure Example
The Korean government has been formulating multiple policies to foster the AI industry, including the National AI Strategy (December 2019), to strengthen global competitiveness. The strategy encompasses 3 key areas, 9 strategic pillars, and 100 execution tasks, mobilizing nationwide capabilities to realize Korea’s AI-driven future vision. The government aims to create up to 455 trillion KRW in economic value by 2030, elevate the nation’s quality of life to the global top 10, and drive paradigm shifts across all industrial and social sectors. In addition, it plans to deploy 3,000 AI-based smart factories by 2030, supporting innovation in manufacturing and SMEs through active and sustained policy support.
#MachineLearning #DeepLearning #AutoML #MLOps #AIModeling
Artificial intelligence (AI) is a technology that artificially implements human learning, reasoning, and perception capabilities—essentially enabling computers to act like a human brain. Traditional machine learning relies on skilled developers to design models that learn based on provided data. However, advancements in technology have made it possible for AI to autonomously create the optimal AI model (AutoML) and automatically manage the generated model (MLOps). The development of these two technologies is rapidly accelerating the sophistication of AI development.
[Fig] Status of AI Adoption and Application Fields
An AutoML solution is “AI that creates AI.” AutoML autonomously extracts data features, selects suitable algorithms, performs learning, and evaluates the resulting models to generate the AI model best suited for the target task. Specifically, it can automatically perform data design, preprocessing, feature identification, feature engineering, algorithm selection, hyperparameter optimization (HPO), model evaluation, and generate improved AI models without manual intervention.
[Fig] Overview of AutoML Technology
An MLOps solution is “AI that manages AI.” MLOps combines machine learning (ML) and operations (Ops) into one integrated system. It provides the management framework necessary to transition conventional research-stage ML models into production-ready ML models, simplifying processes and ensuring maintenance. MLOps automates model maintenance, retraining, deployment, and monitoring—enabling continuous improvement and operational stability of AI systems.
[Fig] Overview of MLOps Technology
Korean MLOps company DeepAuto Co., Ltd. (CEOs: Sungjoo Hwang and Wonyong Jung) successfully raised seed investment from Naver D2SF (D2 Startup Factory). DeepAuto’s MLOps solution automates the full AI lifecycle—data preprocessing, model search and optimization, and deployment—across enterprises adopting AI. In particular, its proprietary AutoMLOps framework has demonstrated superior efficiency compared to big tech solutions. Based on this technological capability, DeepAuto was recently selected for the Deep Tech TIPS program, securing 1.5 billion KRW in funding. The company has already secured partners such as Samsung Advanced Institute of Technology, Cheil Worldwide, and StradVision, and is currently recruiting for its U.S. expansion.
[Fig] DeepAuto Co., Ltd.’s Partner Network
Other Korean companies are also actively developing AutoML and MLOps-related technologies. TEN Co., Ltd. (CEO: Sejin Oh) develops customized MLOps solutions that utilize AI learning data and assetization technology to automatically generate and trade customized learning metadata. The company raised 3 billion KRW in a Pre-A funding round in recognition of this technology. Meanwhile, DSLab Global Co., Ltd. (CEO: Seungki Yeo) has developed auto-labeling technology, which combines automated data generation and labeling techniques to enable even non-experts to develop AI models easily. As such, the development of AutoML and MLOps technologies—AI that creates and manages AI—is progressing actively in Korea.
[Fig] TEN Co., Ltd.’s MLOps Solution Model and DSLab Global’s AutoML Structure Example
The Korean government has been formulating multiple policies to foster the AI industry, including the National AI Strategy (December 2019), to strengthen global competitiveness. The strategy encompasses 3 key areas, 9 strategic pillars, and 100 execution tasks, mobilizing nationwide capabilities to realize Korea’s AI-driven future vision. The government aims to create up to 455 trillion KRW in economic value by 2030, elevate the nation’s quality of life to the global top 10, and drive paradigm shifts across all industrial and social sectors. In addition, it plans to deploy 3,000 AI-based smart factories by 2030, supporting innovation in manufacturing and SMEs through active and sustained policy support.
#MachineLearning #DeepLearning #AutoML #MLOps #AIModeling