Track Overview
At PyTorch Conference 2025, tracks are designed around core themes related to AI technology, research, applications and PyTorch Foundation projects –helping you focus on the topics most relevant to your interests and expertise. 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!
Agentic Data
Explore the future of intelligent data pipelines that shape, curate, and enhance AI training.
Who should attend?
Data scientists, ML engineers, researchers.
What will you learn?
Tools and methods for smarter data pipelines and adaptive datasets.
Deep Learning Compilers, Kernel Authoring & Accelerators
Go deep into PyTorch extensibility with Deep Learning Compilers, custom kernels, and hardware acceleration.
Who should attend?
Developers, systems engineers, hardware specialists.
What will you learn?
How to extend PyTorch, optimize kernels, and maximize accelerator performance.
EdgE
Discover how PyTorch powers AI on mobile, embedded, and IoT devices.
Who should attend?
Engineers building edge, mobile, and IoT AI solutions.
What will you learn?
Deployment strategies, model compression, and on-device optimization.
Foundation Projects / Ecosystem
Get the latest on vLLM, Deepspeed platform projects as well as ecosystem projects from Huggingface, Lightning, and all our community.
Who should attend?
PyTorch users, contributors, and ecosystem developers.
What will you learn?
Updates, roadmaps, and integration best practices for PyTorch ecosystem tools.
LLMs
Dive into the latest breakthroughs in Large Language Models with PyTorch.
Who should attend?
Researchers, ML engineers, developers, product teams.
What will you learn?
Architectures, training optimizations, and real-world LLM applications.
PyTorch at Scale
Learn how to build, train, and serve models at massive scale.
Who should attend?
ML practitioners and infrastructure engineers.
What will you learn?
Distributed training, parallelism, and scaling AI in production.
PyTorch in Academia
See how PyTorch is advancing research and education across disciplines.
Who should attend?
Researchers, students, educators.
What will you learn?
Cutting-edge research, teaching practices, and academic collaboration with PyTorch.
PyTorch in Industry
Explore real-world use cases of PyTorch driving business innovation.
Who should attend?
Industry practitioners, product leaders, technical decision-makers.
What will you learn?
Deployment strategies, lessons from production, and industry success stories.
Reinforcement Learning
Discover the latest in RL algorithms, frameworks, and real-world applications.
Who should attend?
Researchers and engineers exploring RL.
What will you learn?
RL techniques in PyTorch for robotics, gaming, and optimization.
🪧 Poster Sessions
Poster Sessions 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 digital posters.
Who should attend?
Poster Sessions 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 Sessions 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 sessions encourage interactive engagement, allowing participants to ask questions, exchange insights, and deepen their understanding of specific areas of interest within the PyTorch community.
