Embedded Files
CODAI 2027
CODAI 2027
The Annual Meeting of the AI Compiler Community
The Annual Meeting of the AI Compiler Community
The CODAI Mission
The CODAI Mission
The goal of CODAI 2027 is to bring together the emerging energy in the AI compiler communities and AI accelerator communities in both academic and industrial research. These realms of research have the opportunity to deliver a pervasive and seamless end-to-end tooling that connects hardware and software development methodologies. This event mainly focuses on: CODAI spans the full spectrum, from ultra-low-power edge devices to datacenter-scale LLM training and inference.
Community
Community
Bring together the emerging energy in the AI compiler communities and AI accelerator communities in both academic and industrial research.
LLM Deployment
LLM Deployment
Discussing new impulses for deploying LLMs and other neural networks, from edge devices to datacenter inference.
Efficient AI, Edge to Cloud
Efficient AI, Edge to Cloud
Fostering, developing, and exploring techniques such as compression, quantization, sparsity, and virtual prototyping to optimize AI from microcontrollers to datacenter accelerators.
Real-World AI Solutions
Real-World AI Solutions
Transforming AI research into impactful industry solution
Topics of Interest
Topics of Interest
MLIR, LLVM, TVM, XLA, Triton, and other compiler frameworks
Kernel generation and autotuning
LLM inference and serving (attention, KV-cache, batching)
Distributed training and collectives
Low-precision formats (FP8/FP4), quantization, sparsity
Heterogeneous and multi-chip compilation
Edge and embedded AI deployment
Testimonials
Testimonials
Prof. Dr. Fredrik Kjolstad
Prof. Dr. Fredrik Kjolstad
Stanford University, USA
"CODAI offered a dynamic forum for exploring AI compiler and accelerator developments for AI applications. Sharing insights on 'Software and Hardware for Sparse ML' was rewarding, contributing to the workshop's goal of fostering seamless hardware-software integration."
Prof. Dr.-Ing.
Jeronimo Castrillon
Prof. Dr.-Ing.
Jeronimo Castrillon
TU Dresden, Germany
"CODAI was an outstanding event, with top contributions exploring the intersection of AI compilers and Edge AI. It was a pleasure to contribute insights on 'Next-generation Compilers for Emerging Systems' and I truly enjoyed the follow-up discussions on innovative tools to effectively bridge software and hardware.”
Prof. Dr.-Ing. Daniel
Mueller-Gritschneder
Prof. Dr.-Ing. Daniel
Mueller-Gritschneder
TU Wien, Austria
"An often overlooked but major factor for the success of a new AI platform is the availability of a mature software toolchain and AI compiler. Participating in CODAI was, hence, a highly rewarding experience, delving into the latest advancements in AI compilers and accelerators tailored for AI applications. It was a privilege to engage in the workshop's lively discussions. CODAI offers a unique collaborative and focused environment for exchanging knowledge to address the AI toolchain challenges."
Prof. Dr. Oliver Bringmann
Prof. Dr. Oliver Bringmann
University of Tübingen, Germany
"CODAI provided a unique platform for presenting the latest developments in AI compilers and accelerators for AI platforms to bridge the gap between the rapid advances in machine learning and efficient hardware implementation. It was a great experience to contribute to the vibrant discussions and collaborative atmosphere of the workshop."
Contact
Contact
For questions contact:
Dr. Michael J. Klaiber, EnCharge AI
info (a) codai-workshop.com
Page updated
Google Sites
Report abuse