GPU, CUDA, and PyTorch Performance Optimizations

Description

This monthly online meetup brings together engineers, researchers, and AI practitioners to explore cutting-edge GPU, CUDA, and PyTorch performance optimization techniques. Hosted by Chris Fregly and Antje Barth, each session combines practical insights with deep technical discussions on maximizing computational efficiency in AI workloads. ## What to Expect The August 2026 session features two in-depth talks focused on GPU, PyTorch, and CUDA performance optimizations. Whether you're building machine learning models, training large neural networks, or optimizing inference pipelines, you'll gain actionable techniques to reduce latency, improve throughput, and make better use of GPU resources. ## Who Should Attend This meetup is ideal for: - Machine learning engineers working with PyTorch and GPU-accelerated workflows - Performance engineers optimizing AI inference and training pipelines - Researchers exploring CUDA programming and low-level GPU optimizations - DevOps professionals managing AI infrastructure and resource utilization - Anyone curious about squeezing maximum performance from their AI systems ## Format The event runs entirely online via Zoom, making it accessible to the global AI community. Sessions begin with introductions and meetup updates from Chris Fregly and Antje Barth, followed by technical presentations and interactive Q&A. The monthly cadence (every 3rd Monday) creates a consistent learning rhythm for participants tracking the latest performance engineering developments. ## Additional Resources Between meetups, participants can dive deeper through the accompanying [GitHub repository](http://github.com/cfregly/ai-performance-engineering/) featuring code examples and optimization patterns, the O'Reilly book ["AI Systems Performance Engineering"](https://www.amazon.com/Systems-Performance-Engineering-Optimizing-Algorithms/dp/B0F47689K8/), the [AI Performance Engineering YouTube channel](https://www.youtube.com/@AIPerformanceEngineering), and a free Generative AI course on [DeepLearning.AI](https://bit.ly/gllm). Whether you're troubleshooting slow training runs, optimizing production inference, or exploring low-level CUDA kernels, this community provides the technical depth and practical experience to level up your AI performance engineering skills. [Join the meetup and get the Zoom link](https://www.meetup.com/ai-performance-engineering-meetup-madrid/events/ckvnttyjclbwb/?eventOrigin=group_upcoming_events) to connect with fellow p