发表机构
Google DeepMind; Google; MIT（谷歌DeepMind; 谷歌; 麻省理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出Coco，一个面向TPU架构师的智能体平台，通过四层架构（数据存储、工具库、智能体、UX）加速硬件-软件协同设计生命周期，减少模拟与洞察时间，并论证协同设计是独特的智能体领域。
AI 中文摘要
协同设计机器学习模型及其运行的加速器是一种不寻常的推理任务：架构师必须对尚不存在的系统做出自信且高风险的结论，而模型演化和硬件节奏的速度意味着分析负担每个季度都在增长。每个决策背后的证据——针对新颖设计点的数百GB全新模拟扫描——在构造上必然缺失于任何LLM的预训练语料库中，且没有外部文献可检索；朴素的“与数据聊天”方法在正确性最关键的地方恰恰会产生幻觉。我们提出Coco（协同设计副驾驶），一个与TPU架构师部署的智能体平台，加速了设置实验、扫描模拟器和获取洞察的协同设计生命周期。Coco构建为四层：（i）数据存储，自动将每次模拟扫描注册到规范化的关系模式中，使智能体将每个数字基于SQL查询而非抓取异构文件；（ii）具有类型化API的工具库，智能体无需人工编排即可组合；（iii）编码重复分析工作流的智能体——最值得注意的是等执行分析，它比较在匹配执行配置下的系统，包括在Pareto前沿之外被扫描但占优的点；（iv）平台UX，其导航状态兼作智能体上下文。我们报告了早期部署经验，朝着减少模拟时间和洞察时间的方向，并论证协同设计是一个独特的智能体领域：其数据必须被检索而非记忆，其工作流是重复但依赖上下文的，专家采用取决于平衡IDE风格控制与交互式探索的UX。
英文摘要
Co-designing ML models and the accelerators that run them is an unusual reasoning task: architects must draw confident, high-stakes conclusions about systems that do not yet exist, and the pace of both model evolution and hardware cadence means the analysis burden grows every quarter. The evidence behind each decision--hundreds of gigabytes of fresh simulation sweeps over novel design points--is by construction absent from any LLM's pretraining corpus, and there is no external literature to retrieve; naive "chat-with-your-data" approaches hallucinate exactly where correctness matters most. We present Coco (Copilot for Codesign), an agentic platform deployed with TPU architects that accelerates the co-design lifecycle of setting up experiments, sweeping simulators, and deriving insights. Coco is built as four layers: (i) a datastore that automatically registers every simulation sweep into a normalized relational schema, so agents ground every number in a SQL query rather than scraping heterogeneous files; (ii) a library of tools with typed APIs that agents compose without human orchestration; (iii) agents that encode recurring analysis workflows--most notably iso-execution analysis, which compares systems at matched execution configurations, including swept-but-dominated points off the Pareto frontier; and (iv) a platform UX whose navigation state doubles as agent context. We report early deployment experience toward a reduction in time-to-simulation and time-to-insight, and argue that co-design is a distinct agentic domain: its data must be retrieved rather than memorized, its workflows are recurring but context-dependent, and expert adoption hinges on UX that balances IDE-style control with interactive exploration.
