DevConnectDevConnect
Sign up · Log in
← back to the feed
1

CUDA - kernel

CUDA Learning Repository This repository is my hands‑on lab for learning CUDA, GPU programming, and parallel computing on NVIDIA GPUs. I’m using it to: - Practice writing and profiling CUDA kernels - Implement classic parallel algorithms (vector ops, matrix ops, reductions, etc.) - Experiment with memory patterns, shared memory, and basic optimizations - Build a public portfolio in systems + AI / HPC The code here is primarily educational: small examples, mini‑projects, and notes as I go deeper into GPU acceleration for high‑performance and deep‑learning workloads. If you work with CUDA, HPC, or performance‑critical code, feedback, issues, and suggestions are welcome. Repository: CUDA
Add a comment
0/2000