Software Engineer building large-scale data infrastructure and LLM-powered AI agents.

Currently working at SK Planet on the AI Data Platform team, where I manage enterprise-grade Hadoop ecosystems and develop internal data platform tools that serve the entire organization.

• Managing 1,900+ Hive tables across 48 databases on a large-scale Hadoop cluster
• Leading Hadoop 2 → 3 migration for enterprise data infrastructure

• Designing and implementing data access control architectures with Apache Ranger
• Ensuring secure, governed access to data assets across the organization

• Developing an LLM-powered Text-to-SQL Agent providing a natural language interface for complex Hive/HiveQL environments.
• Analyzing complex BI-generated queries and building training datasets
• Bridging the gap between natural language and enterprise data queries

Data & Distributed Systems
Hadoop HDFS Hive HiveQL Trino Apache Ranger ClickHouse

Backend & DevOps
Python Node.js Express Nginx PM2 Linux Shell Script

Frontend & Full-Stack
React MongoDB

Monitoring & Testing
Langfuse Locust Grafana

Package & Process Management
uv pip npm Docker

• langfuse/langfuse-python - Open-source LLM observability & analytics
  • Contributed to [fix: apply stricter early routing for base64 media to prevent SSE dat… ] (#1544)

• Open Source: Actively contributing to Python-based AI & data infrastructure projects.
• Optimization: Infrastructure tuning and performance engineering for large-scale Hadoop clusters.
• Knowledge Sharing: Technical blogging and building tools to boost data team productivity.