Writing code is the baseline. The real value lies in seeing the system as a living organism—optimizing for liquidity, scalability, and tangible business velocity.
In an era where AI writes code, an engineer's value shifts from execution to architecture. I specialize in full-stack development built on scalable serverless infrastructure—keeping overhead low while maximizing business impact.
As AI agents become better at using tools, the bottleneck shifts from model quality to permission design. This post lays out a least-privilege operating model for MCP-based tool use.
A practical explanation of vector databases, how they differ from traditional databases, and why they matter for AI search, RAG, and recommendation systems.
A technical explanation of how vector databases store embeddings, search them with ANN, and match results back to original documents.
Turn fragmented enterprise data into a living knowledge asset.

A smart parenting app that combines daily logs, AI analysis, local play maps, and community support

Event-driven serverless system for large-scale genomic data processing
AI-powered EMR system for Korean medicine clinics

