seon99il

About me

Hi, I’m Seonggil Jeong.

A data engineer who takes responsibility for data quality. I focus on making data that anyone can trust, whatever they use it for.

Visible data vs. trustworthy data

Data showing up on a screen and data being trustworthy are two different problems. A number can always be made to look plausible, but if no one can explain where it came from and what it went through, it is a weak basis for a decision. I want to be the person who stands behind data, from its origin to its quality, before it reaches anyone’s hands.

What I’m most wary of

What worries me most is not the error itself but not knowing an error happened. Data goes wrong everywhere, but when loss or reordering goes unnoticed, every analysis built on top is quietly contaminated. That’s why I value surfacing problems as much as preventing them.

What I look at

  • Lineage: can I trace where a value came from and what it passed through?
  • Blast radius: when something breaks, how far does it spread, and can it be stopped before it does?
  • Observability: when quality breaks, does a person find out first, or the user?
  • Safe re-runs: does running the same job again give the same result (idempotency)?
  • Eventual consistency: delays are acceptable, but does the data always converge to the same value?

What I’m aiming for

I’m drawn to teams that guarantee data quality through structure, not individual carefulness. Validation should be part of the pipeline, and trust should be something you can demonstrate, not just claim. In that kind of environment, I want to be an engineer who can say with confidence: “you can rely on this data.”

This blog is where I write down what I learn along the way.