Systems I Am Thinking About
A short note on the kind of infrastructure problems I like: databases, search, concurrency, and correctness.
I like software that has to be correct while the world is actively trying to make it weird.
Distributed databases, search engines, schema migration tools, scheduling systems - all of them live in that zone where the happy path is only the opening act. The fun part is asking what happens when clocks drift, requests overlap, users scroll too fast, caches stampede, intervals collide, schemas branch, or a supposedly clean abstraction starts leaking timing and state.
That is the kind of work I keep getting pulled toward.
What I Like Building
I like backend systems where the behavior can be explained without shrugging.
- Search systems with predictable latency and a ranking pipeline that makes sense
- Database tooling that treats history, migrations, and conflicts as first-class things
- Distributed systems that prefer explicit coordination over vibes
- Performance fixes where the real win is deleting accidental quadratic work
- Security and verification work where every claim says exactly what it proves
The common thread is not a particular stack. It is the feeling of turning a messy system into something calmer.
Current Direction
Right now I am especially interested in realtime databases, distributed coordination, indexing, schema versioning, and correctness tools for infrastructure.
I am also trying to write more. Systems work can look intimidating from the outside, but a lot of it becomes much friendlier when someone slows down and explains the invariants, tradeoffs, and failure modes plainly.
That is the energy I want this little reading/writing corner to have.