Kakao Data Information Platform Team: Platformizing the Entire Process from Data Collection to ML/AI Utilization

The Kakao Data Information Platform Team platformizes the entire process from data collection to information extraction and provides it as a SaaS to enable Kakao crew members to handle data easily. The team develops and operates various services that automate repetitive tasks and facilitate data-driven decision-making.

Key Service Areas

The platform operated by the team is structured according to the data lifecycle as follows:

• Data Collection and Storage: Collects data via Data Collection SDK and Server, and processes and stores it using Realtime/Batch Process Pipeline and Data Lake.
• Data Analysis and Visualization: Supports data analysis through Data Analytics, Self Service BI, Analytics Notebook Service, etc.
• Data Management: Strengthens data governance with Data Discovery, Access Control, Quality Monitoring, and Lineage Service.
• Data Utilization and Experimentation: Assists with data utilization and experimentation through Self Service Data Preparation, AB Test, Realtime Data Pipeline, Data Query, MLOps Platform Service, etc.

Tech Stack and Work Culture

For platform service development, we primarily use Python, Scala, Kotlin, Java, and apply various technologies including DevOps (k8s, argocd), real-time pipelines (kafka, flink), batch pipelines (airflow), big data (hadoop, spark, trino), and storage (mysql, redis).

Our work style prioritizes asynchronous communication optimized for remote environments, establishing a culture of documenting and discussing in writing. We emphasize data-centric thinking—validating hypotheses and measuring improvements with data—and the initiative to solve problems within the team as key competencies.