Algorix Raises Pre-Seed Funding From Kakao Ventures to Build Unified Data Layer for Enterprise AI

South Korean AI data infrastructure startup Algorix has secured pre-seed funding from Kakao Ventures.

Algorix is building a data engine that unifies structured and unstructured multimodal data scattered across a company’s internal systems into a single logical layer. The platform automates a process that has traditionally required people to manually organize and connect disparate data formats — text, tables, images, and relational information — so that AI can query, process, and analyze them consistently without switching between separate systems for each data type. Rather than treating scattered data as something to simply search, Algorix turns it into a layer that AI can read, write, and compute against directly within real business workflows.

As AI shifts from answering simple questions to acting as an agentic system that autonomously navigates systems and data on its own, data fragmentation inside enterprises has emerged as a major bottleneck. Structured and unstructured data typically sit in separate stacks across different systems, forcing companies to build custom configurations just to work with them together. Even when the information a company needs already exists internally, someone still has to manually extract and connect it. While heavy investment has already flowed into document recognition and data analytics, the step in between — connecting and structuring fragmented data — has remained a largely unautomated gap. Algorix is positioning its multimodal data integration layer to fill that gap.

With the new funding, Algorix plans to accelerate both product development and go-to-market efforts. Leveraging its capabilities in designing scalable, cloud-based data platforms, the company is currently running proof-of-concept projects with multiple enterprise clients. It is also planning a software partnership with AWS that would let customers deploy the platform directly within their own cloud environments. Algorix intends to first target verticals with highly sensitive data before expanding into the APAC and North American markets over the longer term.

The team is led by CEO Donghan Kwon, who saw the structural limits of enterprise data integration firsthand while running AI transformation projects for large corporations, alongside Chief Scientist Suho Noh, who brings broad research and development experience spanning LLM agents, machine learning, database systems, and backend engineering. The founders say that simply assembling good tools doesn’t produce a system that actually works, which is why they are focused on solving data integration as a core problem — building the multimodal data layer they see as a prerequisite for any AI-native workplace.

“Algorix is tackling a problem that’s essential to how companies actually get work done — turning the unstructured knowledge scattered across an organization into a structured system that AI can trust and use,” said Hyunik Cho, senior investment associate at Kakao Ventures. “We expect them to build core data infrastructure for enterprise AI transformation and lead the shift toward AI-native workplaces.”

“As AI transformation accelerates, every system built on top of it ultimately comes down to what data layer it runs on,” said Donghan Kwon, CEO of Algorix. “We’re not aiming to be a single product — we want to build the infrastructure layer that underlies every workflow in which AI handles data.”