Emre's Blog
emre570.bearblog.dev
A personal blog by Emre, a machine learning systems and performance engineer, exploring inference optimization, low-latency audio models, and CUDA/CuTe programming.
The blog shares practical techniques for building and speeding up ML inference systems, with a focus on audio models and GPU kernels. Posts draw from production experience, offering concrete examples in CUDA and CuTe rather than abstract theory. Engineers working on low-level performance will find applied insights here.
Checked with Google Web Risk · 1 day ago
- language
- EN
- here since
- July 2026