Datumo - a startup based in Seoul, started with AI data labeling services, now developing tools to help businesses check, monitor and improve AI models without technical expertise. Recently, Datumo announced the successful mobilization of 15.5 million USD, raising the total raised capital to about 28 million USD from funds such as: Salesforce Ventures, KB Investment, ACVC Partners, SBI Investment and a number of other investors.
CEO David Kim - a former AI researcher at the Korea Defense Development Agency, came up with the idea of labeling manual data that is too time-consuming. In 2018, he and 5 former KAIST founded Datumo (formerly known as SelectStar). The company quickly achieved a revenue of more than 1 million USD in the first year and served large corporations such as Samsung, LG, Hyundai, Naver, SK Telecom. In 2024, revenue will reach about 6 million USD with more than 300 customers.
In recent years, Datumo has expanded to evaluate AI models and release the first standard data set in Korea on AI reliability and safety. The company provides Datumo Eval platform - an evaluation tool that does not require writing code, automatically generating testing data to detect unsafe, biased or incorrect feedback.
Datumo is different from competitors such as Scale AI, Galileo or Arize AI by owning a licensed data set from published books, which is highly arguative but difficult to handle. The new capital will be used to promote R&D, develop AI assessment tools for businesses, and expand marketing in Korea, Japan and the US. Datumo currently has 150 employees in Seoul and opened an office in Silicon Valley in March.
