This blog is a public record of how I think: paper reviews, failure analysis, and applied ML systems lessons.
What you can expect
- Paper reviews with assumptions → failure modes → what I’d change
- Reproducible experiments when possible
- Notes aimed at senior/staff-level ML work