1
原《星空》首席系统设计师库尔曼：可能不会再有《星空 2》了
IT之家 9 月 24 日消息，据外媒 VGC 今天（24 日）晚间报道，曾担任《星空》首席系统设计师的库尔特 · 库尔曼认为，《星空》恐怕不会迎来续作，这个系列也可能因此无法完全实现原本的潜力。 当被问及 B 社未来是否会考虑开发新的《星空》作品时，库尔曼给出了否定的判断。“如果真的推出续作，我会…
2
LLM Tool Failures: Only 3 Root Causes – Value, Condition, Intent
大模型工具调用总翻车？三大根因一次讲透，从“填表人”视角避开90%的坑。
Article URL: https://github.com/Jang-woo-AnnaSoft/execution-state-preflight/blob/main/who-fills-in-the-form.md Comments URL: https://news.ycombinator.…
3
How to Quickly Understand and Improve AI Agent Design
从系统设计视角拆解AI Agent构建，借Bob大叔观点快速提升设计敏感度，直击编码Agent的软肋。
Coding with agents really has surfaced a bunch of problems and challenges we never ran into when writing code by hand. Agents work even faster than yo…
4
Lucid Train
还在为接手老代码发愁？Lucid Train帮你快速构建系统设计，理清代码库架构，开发效率直接翻倍！
Build system design for new and existing codebase Discussion | Link
5
Software Engineering for AI-driven Building Operation
AI落地建筑运营的软件工程指南，涵盖系统设计到部署的完整链路。
arXiv:2608.16237v1 Announce Type: cross Abstract: Building operations are energy-inefficient. Artificial Intelligence (AI)-driven control systems prom…
6
The case for disaggregated LLM serving
把大模型推理拆成预填充和解码两段，用网络传KV缓存，破解长上下文延迟痛点，值得一读。
Article URL: https://blog.doubleword.ai/when-to-disaggregate Comments URL: https://news.ycombinator.com/item?id=49269470 Points: 4 # Comments: 0
7
What Happens When 20,000 People Click the Same Seat
万人同时抢同一个座位，票务系统如何避免超卖？真实压测经验揭秘
Building a seat picker is easy. You render a map, you colour the free seats blue, you let people click. Then tickets go on sale for a big fixture, and…
8
Failure-Aware Long-Form Translation: Design and Implementation of a Recoverable LLM Translation System
长文本翻译总在长句中崩盘？这套可恢复的LLM翻译系统设计，让失败不再重来。
arXiv:2608.09187v1 Announce Type: new Abstract: A long-form translation request can succeed at the API layer and still produce an unusable result. The…
9
A Control System, a Dataset, and a Recipe for Making Frozen LLM Agents Learn a Domain
冻结LLM代理如何通过控制系统与数据集学习特定领域？这篇论文给出了系统方案。
arXiv:2607.25415v1 Announce Type: new Abstract: Production LLM agents are increasingly assembled from a frozen model wrapped in a harness: a prompt te…
10
Designing Safety-Constrained LLM Systems for Public Health Information Access
为公共卫生信息获取设计安全约束大模型系统，聚焦母婴健康资源导航
arXiv:2607.13038v1 Announce Type: cross Abstract: We present the design and implementation of a safety constrained large language model (LLM) system f…
11
Sangam: Efficiently Serving Diffusion LLMs with the AR Stack
扩散LLM推理效率新突破：Sangam用AR Stack化解双向注意力缓存难题，大幅加速生成服务。
arXiv:2607.04206v1 Announce Type: cross Abstract: Diffusion language models (dLLMs) generate text by iteratively denoising a masked response and can c…
12
AI Workflows Need Topological Sort
AI工作流依赖顺序执行，拓扑排序能避免依赖冲突，提升管线稳定性。
Article URL: https://arpitbhayani.me/blogs/ai-topological-sort/ Comments URL: https://news.ycombinator.com/item?id=48797087 Points: 1 # Comments: 0
13
Show HN: Rethinking Anti-Cheat: Can Trust Be Distributed?
把反作弊从“中心化监控”转向“分布式信任”，游戏安全的新思路值得一看。
Article URL: https://medium.com/@www.ahadprogamer/rethinking-anti-cheat-can-trust-be-distributed-b37a353ec84b Comments URL: https://news.ycombinator.c…
14
The Shop on the Corner: How I Learned System Design Without Building Clone
街角小店隐喻系统设计，从柜台到API，轻松理解高可用架构
The Shop on the Corner: How I Learned System Design Without Building Amazon Nephew asks his uncle — 10 years deep into building large-scale systems — …
15
Multi-Region Architecture: Going Global Without Going Broke
业务走向全球，多区域架构如何兼顾性能与成本？看这篇就够了。
When an application grows geographically, it is logical to start serving it from a second region to improve latency and availability.
16
Context vs. Memory Engineering in Agentic AI Systems
区分上下文工程与记忆工程，掌握智能体AI推理与记忆的关键差异，提升系统设计能力。
Article URL: https://machinelearningmastery.com/context-vs-memory-engineering-in-agentic-ai-systems/ Comments URL: https://news.ycombinator.com/item?i…
17
Unhappy Path Changed My Backend Perspective...
从只盯着正常流程到重视异常路径，这是后端开发者认知升级的关键一课。
One thing that backend slowly taught me... When I started learning backend, my focus was always on the happy path. I used to think, user sends the cor…
18
AI's Event Backbone
揭秘AI系统中的事件基石，如何用事件驱动构建更灵活的AI架构？
Article URL: https://altertable.ai/blog/2026-03-10-ai-s-events-backbone Comments URL: https://news.ycombinator.com/item?id=48751604 Points: 1 # Commen…
19
Collective cooperation without individual fidelity in LLM agents
一项新研究揭示大模型智能体无需个体忠诚也能实现集体协作，对多智能体系统设计有启发。
arXiv:2606.30454v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as agents in simulations of social systems, yet i…
20
Signal: A draft UX framework for LLM-based systems
专为LLM系统设计的UX框架草案，从交互体验角度优化AI应用设计
Article URL: https://github.com/hi-mundo/SIGNAL Comments URL: https://news.ycombinator.com/item?id=48746626 Points: 1 # Comments: 0