High-performance GPU kernels for LLM inference in OpenAI Triton. Fused RMSNorm, SwiGLU, INT8 GEMM with benchmarks and roofline analysis.

Unified benchmarking and profiling framework for the JAX scientific ML ecosystem. Timing, GPU/energy monitoring, FLOPS counting, roofline analysis, statistical testing, regression detection, and CI integration.

ARM NEON SIMD optimizations on Apple Silicon for attention (31.88 GFLOPS/s, 11.1x speedup) and Black-Scholes (1.03x, proving SIMD fails on transcendental-heavy workloads). Roofline analysis predicts optimization potential before implementation. OpenMP parallelization.

Kernel-level profiling of batch-1 decode on consumer GPU to prove that decode is memory-bandwidth-bound against the hardware roofline, then beating the baseline with a fused dequant+GEMV CUDA kernel. (regime-aware attribution at the end)

High-performance computing project exploring CPU and GPU matrix multiplication optimization through OpenMP, CUDA, tiled kernels, and cuBLAS, with detailed performance analysis across large matrix sizes

Three measured notebooks on GPU inference optimization: roofline analysis, KV-cache decode optimization (4.21x), and CUDA-graph launch-overhead elimination (5.38x). Pure PyTorch.