FIELD NOTES / IN WRITING
A place for questions and notes
Research questions, engineering decisions, and small things worth keeping along the way.
When I work on research or software, the final code and results tell only part of the story. Why did the work begin with this question? Why did an approach stop being useful? Which small change helped something make sense? I want to keep a record of those parts, too.
That is what this space is for.
What I will write about
My interests include AI, code security, and how models and systems behave on practical problems. The writing here will gradually grow around three areas:
- Research notes: questions, experimental choices, and results in context, including what remains unexplained.
- Project records: the problem behind a project, its implementation and trade-offs, with links to material that is available to share.
- Development and learning: technical problems, experiences with tools, and notes I may want to return to later.
The first research project featured here is Cyber-JEV. It studies how matched examples from security fixes can inform source-code vulnerability triage. For an introduction to the question and approach, start with the reading guide.
Leaving room for revision
I want the writing here to make clear what was observed, how I interpret it, and what still needs to be tested. An experiment does not always produce a neat result. A limitation described carefully can still be a useful starting point.
When I find a mistake or reach a better understanding, I will update the relevant article and record its revision date.
Beyond research
I also enjoy anime and related works, and I want this site to retain a little of that personal side. The stories and characters I like are part of life alongside papers and code.
Thanks for visiting. I hope this becomes a place to browse, exchange ideas, and find a thread of thought again.