GPU, CUDA, and PyTorch Performance Optimizations

Description

Explore GPU, CUDA and PyTorch performance optimisation in an online technical masterclass listed for [Lagos](https://www.createwith.com/cities/lagos) on 17 August 2026. The session centres on improving the performance of GPU-accelerated workloads and understanding how CUDA and PyTorch fit into that process. ## What to expect This event brings together three closely connected areas of modern AI and high-performance computing: **GPU execution, CUDA programming and PyTorch optimisation**. It offers a focused setting for developers who want to think more critically about runtime performance rather than treating acceleration as a simple hardware upgrade. Based on the published event topic, the discussion will be relevant to questions such as how software makes effective use of GPUs, where CUDA sits within the acceleration stack, and how PyTorch workloads can be approached with performance in mind. The event page does not currently provide a detailed agenda, named speakers or specific demonstrations, so attendees should consult the registration listing for any programme updates. ## Who it is for The technical subject matter makes this event particularly relevant to: - AI and machine learning engineers working with PyTorch - Software developers interested in GPU-accelerated applications - Data scientists seeking a stronger understanding of model runtime performance - Performance engineers investigating CUDA-based workloads - Technical leads evaluating the efficiency of AI systems No formal experience level or prerequisites are stated. However, familiarity with machine learning development, GPU concepts or PyTorch is likely to help attendees get more from a performance-focused session of this kind. ## Event format The event is presented as an **online masterclass and meetup**, making it accessible remotely despite being listed in connection with Lagos. Its focused title suggests a technical session rather than a broad introduction to artificial intelligence, with the emphasis firmly on the engineering and optimisation layer behind computationally intensive workloads.… [Register for this event](https://dev.events/conferences/gpu-cuda-and-py-torch-performance-optimizations-ltspttyjclbwb)