C++23 benchmarking framework with 6 profiler backends, CUDA GPU support, statistical regression detection, cross-compilation for 5 architectures, and CLI tools for analysis and visualization.

Quantum workload planning and profiler-backed architecture analysis for exact tensor-network execution.

CUDA-accelerated kNN regression for rent estimation with CPU baseline, shared-memory optimization, and profiling

16-step CUDA optimization of FlashAttention-2 achieving 99.2% of official performance on A100 — Ampere architecture

High-Performance Computing (HPC) & Optimization studies using CUDA C++. Includes Grid-Stride Loops, Shared Memory tiling, and Nsight Compute profiling analysis.

🎬 Explore GPU training efficiency with FP32 vs FP16 in this modular lab, utilizing Tensor Core acceleration for deep learning insights.

High-performance Sobel edge detection using CUDA with CPU vs GPU benchmarking, roofline analysis, and Nsight profiling.

CUDA C++ practice project for RTX 4070 SUPER — explore GPU concurrency, pinned memory, and Nsight profiling. Includes SAXPY and 2D blur kernels to train optimization, stream overlap, and timing analysis for NVIDIA Developer Technology Engineering skillset.

CUDA Samples and Nsight Guided Profiling Samples

Custom PyTorch CUDA kernel implementing optimized ReLU activation with vectorization, performance profiling, and memory analysis on Tesla T4 GPU achieving 75% bandwidth efficiency.

🚀 High-performance implementations and benchmarks of SSSP and APSP algorithms (Bellman–Ford, Dijkstra, Floyd–Warshall, Johnson) in Serial, OpenMP, CUDA, and Hybrid CPU+GPU. Includes profiling, speedup plots, and HPC notebooks

The MNIST classification problem is a fundamental machine learning task that involves recognizing handwritten digits (0- 9) from a dataset of 70,000 grayscale images (28x28 pixels each). It serves as a benchmark for evaluating machine learning models, particularly neural networks.

A reproducible GPU benchmarking lab that compares FP16 vs FP32 training on MNIST using PyTorch, CuPy, and Nsight profiling tools. This project blends performance engineering with cinematic storytelling—featuring NVTX-tagged training loops, fused CuPy kernels, and a profiler-driven README that narrates the GPU’s inner workings frame by frame.

Accelerate and optimize existing C/C++ CPU-only applications using the most essential CUDA tools and techniques.

University Project for "Computer Architecture" course (MSc Computer Engineering @ University of Pisa). Implementation of a Parallelized Nearest Neighbor Upscaler using CUDA.

learning how to do profiling on a yolov8 net using nvidia nsight compute

Remote development on HPC clusters with VSCode

Matrix multiplication example performed with OpenMP, OpenACC, BLAS, cuBLABS, and CUDA

Fast, reproducible, and portable software development environments