Press Release
Korea Deep Learning launches DEEP Agent, an AI agent that finishes work through documents
Korea Deep LearningPublished

Korea Deep Learning launched DEEP Agent, a document-based AI agent, in March 2025. It is an AI that does not stop at reading documents but finishes the work through them. It interprets the visual structure and the language meaning in a document together to extract the key information, and on that basis takes action on the work, from summarizing a contract to entering it into ERP.
An AI that understands documents and does the next task
DEEP Agent interprets the visual structure and the language meaning in a document at the same time to extract the key information, and takes action on the work based on it. It combines DEEP OCR, the document recognition solution, with DEEP Parser, the document structure analysis solution, and is built on a vision-language model (VLM) that Korea Deep Learning developed in-house.
Contract processing is a typical use. When a contract is uploaded, DEEP Agent analyzes the whole document, summarizes the key clauses, points out the clauses that carry risk, and carries it through to entry into the ERP system. If OCR was a technology for reading text, DEEP Agent is a technology that understands documents and does the task that comes next.
In an interview with ZDNet Korea in February 2025, Kim Ji-hyun, CEO of Korea Deep Learning, said that advances in LLM technology were making it possible to use OCR together with chatbots and agents, and expected this to widen the room for the OCR business to grow. DEEP Agent turns that direction into a product.
Deployed in two weeks on average, with no separate training
Document automation used to require collecting data and training a model for each of a customer's document formats. Built on its own VLM, trained on more than 400 million text and image documents over five years, DEEP Agent is deployed in the field within two weeks on average with no separate training process. Deployment costs and build times also fell. According to the company, the deployment cost dropped from an initial KRW 300 million to KRW 150 million, and the build time from 66 days to 10.
Demand for AI that understands and processes documents
"Demand is growing for AI that goes beyond reading documents to understanding and processing them."
— Kim Ji-hyun, CEO, Korea Deep Learning
About DEEP Agent
Korea Deep Learning makes AI that does not stop at reading documents. DEEP Agent is a document-based AI agent that interprets a document's visual structure and linguistic meaning together to extract the key information, then carries out the whole of the document work, from classification and verification to the hand-off into business systems. Upload a contract and it summarizes the key clauses, flags the risky ones and goes on to register it in the ERP, and it reaches the field in two weeks on average with no separate training. It holds GS Certification Grade 1, Korea's national software quality certification, and is designated for priority purchase by public institutions. More information is available on the DEEP Agent page at koreadeep.com.
About Korea Deep Learning
Korea Deep Learning (KDL), founded in 2019, is an enterprise AI agent company. Built on KDL Frontier, its own vision-language model (VLM) specialized for documents, KDL makes AI agents that take on and complete the document work of businesses and public institutions. Its main products are DEEP OCR, DEEP Parser, DEEP Agent and DEEP Ops.







