Start with the user and failure cost.
I define the decision a user needs to make and validate the riskiest assumption with the smallest useful experiment.
About
I connect ideas to operable services
and document the reasoning.
I study Computer Science at Soongsil University and build backend, data, and AI systems. My implementation boundary includes where data comes from, how a service is deployed, and how failures are observed and recovered.
I prefer defining the problem and constraints before choosing technology. I preserve decisions, rejected alternatives, failed hypotheses, and verified results in code, ADRs, work logs, and troubleshooting records.
How I work
Development principles
Treat operations as implementation.
Deployment, observability, security boundaries, and recovery paths are part of done. Work in progress remains labeled as such.
Keep records verifiable.
I document performance and incidents with reproduction conditions and limitations, without filling gaps with unverified results.
Share decision context.
Small changes, API contracts, ADRs, and tests preserve context so others can understand why a decision changed.