17 Matching Annotations
  1. May 2026
    1. Among some teams at OpenAI, we saw the number of landed PRs increase by 500% in the first three weeks.

      大多数人认为AI辅助编程只能带来适度的生产力提升,但作者认为Symphony系统实现了500%的代码合并增长率,这是一个惊人的数字。这个数据点挑战了人们对AI辅助编程效果的传统预期,表明正确的AI编排可能带来指数级的生产力提升。

    1. Instead of using domain knowledge to prescribe team organization, roles, or workflows, Fugu learns to dynamically assemble agents from a pool and coordinate them through non-obvious but highly efficient collaboration patterns.

      大多数人认为多模型系统需要人工设计明确的分工和角色分配,但作者认为Fugu能够自主发现最优的协作模式。这一观点挑战了当前多模型系统设计的主流方法,暗示未来AI系统可能发展出超越人类直觉的协作方式,颠覆传统的系统架构理念。

  2. Apr 2026
    1. These skills act as an orchestration layer that helps scientists work through broad, ambiguous, and multi-step questions more effectively.

      将AI描述为'编排层'而非简单工具,体现了AI在科学研究中角色的根本转变。这暗示未来科学家可能更像AI系统的指挥者,而非直接执行者,重塑科研工作流程。

  3. Mar 2025
  4. Aug 2024
  5. Dec 2023
  6. Jul 2023
  7. Jun 2023
  8. Nov 2022
  9. Apr 2022
  10. Mar 2022
  11. Nov 2020
    1. Maybe your dbt models depend on source data tables that are populated by Stitch ingest, or by heavy transform jobs running in Spark. Maybe the tables your models build are depended on by analysts building reports in Mode, or ML engineers running experiments using Jupyter notebooks. Whether you’re a full-stack practitioner or a specialized platform team, you’ve probably felt the pain of trying to track dependencies across technologies and concerns. You need an orchestrator.Dagster lets you embed dbt into a wider orchestration graph.

      It can be common for [[data models]] to rely on other sources - where something like [[Dagster]] fits in - is allowing your dbt fit into a wider [[orchestration graph]]

  12. Nov 2019