AI Performance Engineering Meetup (Madrid)

Description

This recurring monthly meetup brings together AI engineers, ML practitioners, and performance optimization specialists in [Madrid](https://www.createwith.com/cities/madrid) for deep technical discussions on GPU acceleration, CUDA programming, and PyTorch performance tuning. Hosted by Chris Fregly and Antje Barth, sessions run online via [Zoom](https://us02web.zoom.us/j/82308186562) every third Monday of the month. ## What You'll Learn Each meetup features practical talks and live demonstrations covering GPU, CUDA, and PyTorch performance optimizations. Topics span the full stack of AI performance engineering — from low-level CUDA kernels to high-level PyTorch model optimizations. Whether you're building production ML systems, training large models, or deploying inference endpoints, you'll gain actionable insights to accelerate your AI workloads. ## Who Should Attend This meetup is ideal for: - ML engineers optimizing training pipelines - AI researchers pushing model performance boundaries - Backend developers working with GPU infrastructure - Performance engineers focused on AI systems - Anyone curious about the intersection of hardware and machine learning Some familiarity with Python and PyTorch is helpful, though sessions cater to various experience levels. ## Format & Resources Sessions begin with introductions and community updates from the hosts, followed by technical presentations exploring specific optimization techniques. The interactive Zoom format allows for real-time Q&A and knowledge sharing among attendees. Between meetups, you can dive deeper through the group's [GitHub repository](http://github.com/cfregly/ai-performance-engineering/), watch previous sessions on [YouTube](https://www.youtube.com/@AIPerformanceEngineering), or explore the hosts' O'Reilly book ["AI Systems Performance Engineering"](https://www.amazon.com/Systems-Performance-Engineering-Optimizing-Algorithms/dp/B0F47689K8/). The group also recommends DeepLearning.ai's [free Generative AI course](https://bit.ly/gllm) for foundational knowledge. ## Why Attend GPU optimization is critical for modern AI development — reducing training costs, improving inference latency, and enabling larger models. This meetup provides practical, production-ready techniques you can apply immediately. Connect with Madrid's AI performance engineering community, share optimization war stories, and stay current with the latest CUDA and PyTorch developments. [Join this recurring meetup series