A beginner's course on CUDA® C/C++
Learn Cuda C programming! This course covers core principles of Cuda C. Master GPU programming for accelerated computing.
Course Content
A beginner's course on CUDA® C/C++ is a beginner-level Udemy course taught by HPC Specialist. It runs 3h 6m, is taught in English, and includes a certificate of completion. Learners rate it 3.7 out of 5 from 91 reviews, with 935 students enrolled. It is a paid course, last updated August 2025.
Looking to unlock the power of NVIDIA GPUs? This course is designed for programmers with a foundation in C or C++ who want to master CUDA C programming. Dive into the fundamentals of CUDA, NVIDIA's proprietary language for general-purpose computing on GPUs (GPGPU).
You'll learn to harness the parallel processing capabilities of NVIDIA GPUs, covering topics applicable to a wide range of CUDA GPUs, from versions 1.0 up to the latest 8.0 and beyond. We'll explore advanced features like dynamic parallelism and warp shuffle functions, introduced in newer CUDA hardware and development toolkits.
Through a blend of engaging lectures and practical example programs, you'll gain the skills to design your own high-performance compute algorithms. Discover how to effectively leverage GPGPU programming to accelerate your applications.
This course also introduces essential profiling and analysis tools, including NVIDIA Nsight Compute. Learn to gain critical insights into your CUDA program's performance, identify bottlenecks, and optimize resource utilization for maximum efficiency.
By the end of this course, you'll have a solid understanding of CUDA C and the ability to develop your own powerful CUDA applications.
- Learn CUDA C fundamentals
- GPGPU Programming with NVIDIA GPUs
- CUDA Profiling using Nsight Compute
- Dynamic Parallelism
- Warp Shuffle Functions
- Develop High-Performance Algorithms
- C/C++ to CUDA Migration
Disclaimer: This course is not affiliated with or endorsed by NVIDIA or its products/services.
Want AI in your product? Hire the team behind TutorialSearch.io
LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.