About
I like building the things other code sits on — databases, queues, rate limiters. The unglamorous layer where the interesting problems live.
Right now that means a vector database in Rust, and whatever the next rabbit hole turns out to be.
Currently — sophomore, B.S. Computer Science
Writing — Rust, Go, C++, Python
Into — storage engines, retrieval, systems that don't fall over
Writing — Rust, Go, C++, Python
Into — storage engines, retrieval, systems that don't fall over
Selected work
From the lab- RUST engramdb Embedded vector database with document chunking and hybrid dense + BM25 retrieval, plus a terminal RAG chat app that runs on local models. ->
- GO go-rate-limiter Per-IP token-bucket rate limiter for HTTP servers, backed by Redis so limits survive restarts and scale across instances. ->
- RUST async-job-queue Asynchronous background job queue exploring task scheduling, retries, and worker coordination. ->
- C++ minidb In-memory database in C++17 with its own command parser and CLI — built to see how databases work from the inside. ->
- PYTHON Dexter CNN that reads the ASL alphabet from a live webcam feed — TensorFlow for training, OpenCV for real-time prediction. ->
- RUST rs-advice A macOS menu bar app that hands you a random piece of advice. Small, useless, delightful. ->
More experiments on GitHub.