Work

Selected areas, projects, and engineering work. This page is intentionally concise and will grow as I publish more project notes.

## Distributed Storage I work on storage systems where metadata scalability, data placement, consistency, failure recovery, caching, and end-to-end I/O performance all matter at the same time. **Focus:** distributed file systems, metadata/data plane design, cache architecture, reliability, C++ and Linux performance. ## AI Infrastructure I am interested in storage infrastructure for AI workloads: feeding GPU clusters efficiently, reducing avoidable data movement, and designing cache and I/O paths around modern training and inference systems. **Focus:** high-throughput access, cache admission and eviction, prefetching, failure handling, GPU-cluster storage paths. ## Systems Engineering A large part of my work is understanding the details below distributed systems: Linux I/O, concurrency primitives, RPC behavior, networking, RDMA, NVMe, profiling, and latency/throughput trade-offs. **Focus:** Linux, C++, concurrency, RPC, RDMA, NVMe, performance analysis. ## Open technical notes I use [Writing](/blog/) for deeper technical notes and essays. Over time, representative projects on this page will link to standalone project write-ups with architecture, trade-offs, and outcomes.