Aug 2026· Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2· pp. 13356-13359· 0 citations· 2 references
Abstract
Writing high-performance system kernels for AI accelerators requires deep hardware expertise and low-level programming skill—a significant barrier for many researchers and engineers. This tutorial presents emerging methods that leverage large language models (LLMs) to automate kernel generation for GPUs and custom accelerators such as AWS Trainium. We cover the full pipeline: supervised fine-tuning on kernel corpora, reinforcement learning-based post-training (PPO, GRPO) with hardware-aware rewards, inference-time scaling for code reasoning, and multi-turn agentic systems for iterative kernel refinement. We also discuss key challenges unique to this domain, including reward hacking in compilation metrics, designing effective evaluation benchmarks for generated kernels, and strategies for data curation across heterogeneous hardware targets. The tutorial bridges the gap between LLM post-training research and systems programming, offering both foundational understanding and practical experience. We demonstrate how recent advances in reinforcement learning and inference-time compute scaling can be adapted to produce correct and performant kernels across diverse hardware backends. Attendees will gain hands-on experience generating Triton and NKI kernels using agentic LLM systems. Tutorial materials are available at https://neuron-science.github.io/llm_kernel_writing.
This work introduces LEAP (Lean Environment-Feedback via Adaptive Pruning), a scalable and computationally efficient multi-turn RL framework optimized for low-level hardware accelerator alignment and proposes a Rank-Based Reward formulation, establishing a practical paradigm for low-level code RL.
The rapid growth in machine learning workloads has fueled the proliferation of custom accelerator architectures. Designed from the ground up, these accelerators often expose programming models that are distinct from GPUs. While hyperscalers and AI chip startups continue to innovate in this space, achieving broad operator coverage to support diverse models remains a major challenge. Additionally, an easy-to-use, high-level kernel programming language is important for rapid iteration of models and kernels. Triton, together with TorchInductor, addresses these issues on GPUs, but its viability on accelerators with different programming models has yet to be established. In this work, we present the first production-scale application of Triton on a custom ML accelerator, MTIA-2i, developed by Meta. To support MTIA-2i, we develop a new compiler backend that targets it, introduce enhancements to TorchInductor code generation, and propose minimal language extensions that expose MTIA-specific architectural features. We demonstrate that Triton-MTIA kernels achieve performance competitive with expert-tuned C++ implementations. Leveraging these development efficiency gains, we successfully deployed manually written and Inductor-generated Triton kernels in production across approximately 60 different model types, accounting for 50% of layers and 47% of non-GEMM execution time for these models. Our results provide compelling evidence that DSLs like Triton can bridge the programming model gaps between ML frameworks, kernels, and custom accelerators, enabling rapid innovation and efficient deployment at scale.
Haishan Zhu, Domi Yan, Michael Levesque-Dion et al.· 0 citations
Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-level expertise. Large Language Models (LLM) can be leveraged together with real-time compiler feedback to build agentic systems for kernel generation. In this work, we present MaxKernel, a multi-agent system that implements three distinct paradigms for TPU kernel development: (1) a Human-in-the-Loop (HITL) agent for collaborative, step-by-step design; (2) an Autonomous (Auto) agent that executes a fully automated, metric/trace-driven optimization loop; and (3) a Graph-Based Autonomous Search that scales the Auto agent for global exploration of the design space. All three paradigms leverage a shared pool of specialized sub-agents to handle planning, implementation, self-debugging, testing, and hardware profiling. We evaluate MaxKernel on JaxBench, a comprehensive suite of 50 diverse kernel tasks for TPUs, alongside complex, real-world workloads from state-of-the-art open-source models. We demonstrate that MaxKernel consistently generates highly optimized implementations, matching expert hand-tuned baselines and delivering significant performance across the benchmark. Our agent is open-sourced and available https://github.com/AI-Hypercomputer/accelerator-agents/tree/main/MaxKernel.
Shang-Kun Wang, Nina Cai, Charles Hoong et al.· 0 citations
We introduce AMDKernelVault, an open HIP and Triton kernel corpus and training framework for recent AMD CDNA GPUs. Existing LLM-based kernel agents are largely CUDA/NVIDIA-centric and often depend on repeated frontier-LLM calls for generation, reflection, and optimization. To address this gap, we develop HIPKernelGen and TritonKernelGen, agent-driven pipelines that transform PyTorch references into HIP or Triton kernels, compile and validate candidates under ROCm, and latency-profile them on AMD hardware. The corpus contains 62,153 execution-verified HIP kernel samples, 2,377 production-grounded ROCm Libraries QA entries, and 39,893 Triton kernels. We further train Qwen3-8B with supervised fine-tuning and execution-aware reinforcement learning as a demonstration of the corpus's utility. Under fixed evaluation budgets, it achieves the highest correctness among the compared models on PyTorch-to-HIP (34.0% Pass@1), TritonBench-G (33.2% Corr@3), and ROCmBench (41.94% Corr@3), but does not uniformly lead compilation or speed metrics. The corpus and documentation are available at https://huggingface.co/datasets/amd/AIG-Datasets, and the associated training and kernel-generation code is available at https://github.com/AMD-AGI/hip_kernel_llm_lab.
Ji Liu, S. Majumder, Yi-Qing Huang et al.· 0 citations
Large language models (LLMs) place unprecedented and still-growing demands on the hardware that trains and serves them. This review surveys the full landscape of AI hardware accelerators for LLMs, including general-purpose GPUs, custom ASICs such as TPUs, Trainium, Groq, and Cerebras, reconfigurable FPGAs, processing-in-memory and near-memory architectures, and emerging neuromorphic and photonic approaches across cloud and edge deployment. Using the transformer's computational structure and roofline analysis as a common framework, we show that the decisive constraint on LLM acceleration is not arithmetic but memory: the autoregressive decode phase is bandwidth-bound, the key-value cache can rival the model weights in size, and data movement dominates energy. Comparing platforms on compute, memory, energy, programmability, and scalability, we find that no single architecture is optimal across workloads: GPUs remain the flexible default and the workhorse of training; domain-specific ASICs win at scale for stable, high-volume workloads; processing-in-memory is the most promising near-term response to the memory wall, entering systems as a heterogeneous complement; and neuromorphic and photonic computing, while promising, are not yet production-ready at frontier scale. Future progress depends on hardware-algorithm co-design and heterogeneous, memory-centric systems: for large language models, the memory system has become the computer.
Tensor algebra workloads, of which deep neural networks are prominent examples, are energy-intensive workloads in modern datacenter and edge deployments, making accelerators necessary to achieve energy efficiency and high throughput. To quickly evaluate and iterate on accelerator designs, we need an accelerator modeling framework that captures salient attributes of devices, circuits, architectures, workloads, as well as optimizing the mapping of the workload onto the hardware. In this paper, we introduce AccelForge, which improves upon existing accelerator modeling frameworks in capabilities, speed, and ease-of-use. AccelForge unifies and multiple works into one framework, and it includes (1) composable user-defined and user-modifiable models of devices, circuits, and architectures, (2) fast mappers that enable accurate evaluation in orders of magnitude less (computer and human) time, and (3) easy-to-use and easy-to-extend, yet still high performance, Python implementations of both the model and mapper to enable rapid research and extension to novel optimizations.
Tanner Andrulis, Michael Gilbert, Vivienne Sze et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
