Track Overview
PyTorch Conference Europe features in-depth technical talks, hands-on workshops, and candid conversations spanning the full AI stack, from bare metal infrastructure to applications and agent-based systems. Each track explores a unique dimension of machine learning and AI—from foundational tools and academic research to large-scale deployment and cutting-edge innovations. With clear tracks, you can easily navigate the program, connect with peers, and deepen your AI knowledge.
Click on a track to learn more!
Agents & Interop
Learn about building autonomous AI agents and ensuring seamless interoperability between different models and systems.
Who should attend?
Platform engineers, AI application developers, and architects designing multi-model systems.
What will you learn?
How to build composable, interoperable agent systems built on top of PyTorch PyTorch, MCP protocols, and agentic architectures.
Applications & Case Studies
Explore real-world successes and practical implementations of PyTorch across various industries, providing insights into deployment challenges and solutions.
Who should attend?
ML practitioners, domain experts, product managers, and anyone looking to see PyTorch applied in real-world scenarios.
What will you learn?
Real-world deployment challenges, domain-specific applications of PyTorch, and practical lessons from production implementations.
Frameworks & Compilers
Gain a deeper understanding of the foundational tools of PyTorch, including optimization techniques, new features in the framework, and advancements in AI compilers for performance.
Who should attend?
PyTorch core developers, compiler engineers, performance engineers, and kernel developers.
What will you learn?
torch.compile internals, kernel authoring, CUDA graph optimization, custom op integration, and compiler-level performance techniques.
GENAI & Multimodal
Discover the latest research and engineering practices in generative AI, including large language models (LLMs), diffusion models, and models that handle multiple data types (text, images, audio).
Who should attend?
Researchers and engineers working with LLMs, diffusion models, multimodal systems, and generative AI applications.
What will you learn?
How to build and optimize generative AI systems spanning video, audio, text, and multimodal applications using PyTorch.
Inference & Production
Master the strategies for deploying and managing models at scale, focusing on low-latency inference, model serving, and MLOps best practices.
Who should attend?
Infrastructure engineers, MLOps teams, and backend engineers deploying and serving models at scale.
What will you learn?
Model serving architectures, inference optimization, KV-cache management, edge/browser deployment, observability, and production scaling strategies.
Responsible AI & Compliance
Understand the principles and tools for developing ethical, fair, and transparent AI systems, along with navigating the evolving landscape of AI regulations and compliance.
Who should attend?
AI governance professionals, policy-minded engineers, product managers, compliance officers, and anyone deploying AI in regulated industries.
What will you learn?
EU AI Act compliance strategies, ethical AI practices, governance frameworks, lineage tracking, and compliance-as-code approaches.
Security & Privacy
Learn about techniques to protect AI models and data from adversarial attacks, ensure data privacy, and secure the AI development pipeline.
Who should attend?
Security engineers, ML engineers concerned with model safety, platform engineers, and anyone deploying AI in security-sensitive environments.
What will you learn?
Threat modeling for AI systems, model serialization security risks, privacy-preserving training, LLM red teaming, and data sovereignty strategies.
Training Systems
Explore innovations in distributed training, hardware acceleration, and system-level optimizations for efficiently training massive and complex AI models.
Who should attend?
ML engineers training large models, distributed systems engineers, and researchers working at scale.
What will you learn?
Distributed training techniques, FP8 precision training, parallelism strategies, debugging distributed systems, and scaling training to thousands of GPUs.
Poster Presentations
Poster Presentations are an interactive showcase where attendees can view digital presentations of projects, use cases, and research related to PyTorch and the broader machine learning ecosystem. Participants will have the chance to engage directly with presenters, ask questions, and exchange ideas around the work displayed in the posters.
Who should attend?
Poster Presentations are ideal for a wide range of attendees, including machine learning practitioners, researchers, engineers, students, and anyone curious about PyTorch applications. Whether you’re looking to explore new research, discover practical implementations, or connect with peers in the community, the Poster Presentations provide a valuable opportunity for learning and networking.
What will you learn?
Attendees can expect to learn about diverse topics within the PyTorch ecosystem, including innovative applications, cutting-edge research, practical tools, best practices, and emerging trends in AI/ML. These presentations encourage interactive engagement, allowing participants to ask questions, exchange insights, and deepen their understanding of specific areas of interest within the PyTorch community.
