1
WEBSITE FOR THE DEV WEEKEND CHALLENGE
用Claude辅助编码，仅周末就做了一个足球主题网站，展示AI如何加速前端原型。
This is a submission for Weekend Challenge: Passion Edition What I Built i build a website about football its legends and the biggest clubs and also n…
2
I'm About to Deploy My Go + React App for the First Time — Wish Me Luck
非开发人员用AI打造20多个内部工具后，首次部署Go+React应用的真实记录，给所有曾被说不会编程的人打气。
Here's the stack: Backend: Go, Gin, JWT auth, PostgreSQL (hosted on Neon.tech), Cloudinary for image uploads Frontend: React + Tailwind CSS I haven't …
3
ASK in the Dark: Uncertainty-Gated LLM Assistance under Partial Observability
利用不确定性门控机制，提升LLM在信息不完全时的高效辅助能力，为AI决策提供新思路。
arXiv:2607.02686v1 Announce Type: new Abstract: Reinforcement learning agents operating under partial observability must act on incomplete information…
4
Three-Phase Evaluation of AI-Assisted Software Development Life Cycle
论文提出三阶段评估框架，系统度量AI在软件开发生命周期各环节的实际效能与局限性
arXiv:2607.05125v1 Announce Type: cross Abstract: This paper presents an exploratory evaluation of how increasing levels of AI autonomy affect softwar…
5
Ask HN: Is anyone experimenting with different ways of using LLMs for coding?
探讨如何突破LLM编程时的频繁中断感，让AI真正成为思维的助力而非阻碍。
I'm a bit annoyed by the feeling that we're kind of stuck when it comes to using LLMs for programming. I use Claude Code and Codex, but I haven't been…
6
I gave my Nuxt Supabase auth scaffold some personality on top of Nuxt UI
用AI辅助设计Nuxt Supabase认证脚手架，按钮动画令人上瘾，配色排版也颇具亮点。
I'm a person that appreciates consistency across applications, especially applications that belong to a solo founder or indie developer alike. I'm not…
7
AI-Assisted Discovery of Convex Relaxations via Dual Agents
双智能体驱动AI，革新凸松弛发现路径，优化理论前沿突破。
arXiv:2606.31182v1 Announce Type: new Abstract: Recent work shows that LLM agents can improve sharp-constant inequalities by searching for extremal co…
8
Ask HN: How do you share /skills across your dev team?
团队如何共享AI辅助开发中的技能配置？.claude/skills的Git管理困境与解决方案探讨。
My team wants to adopt a shared set of skills in our codebase: design rules, reusable component references, copy guidelines, etc. Currently, we've kep…
9
Show HN: Autoharness – a self-learning, maintaining skill layer for Claude Code
为Claude Code注入自学习技能层，让AI助手能持续优化自己的行为模式。
Article URL: https://github.com/tigerless-labs/autoharness Comments URL: https://news.ycombinator.com/item?id=48732795 Points: 2 # Comments: 0
10
顶刊生物实验难复现？统一操作话术来了！编译通过率98.6%
顶刊生物实验难复现？AI统一操作话术，编译通过率98.6%，让科研更规范高效。
AI进入生物制造，真正缺的不是更聪明的大脑，而是连接数字与物理世界的接口。
11
How I stopped rebuilding the same React features from scratch (using Claude Code skills)
用Claude Code技能为React项目按需自动激活，告别重复造轮子
Every React project starts the same way. You open a new repo, and the first thing you need is auth. So you ask Claude Code to build it. And it generat…
12
What GPT Told Me About My Paper — And What I Actually Learned
用GPT帮你读论文，N=1的工程案例竟能学到这么多反思。
What GPT Told Me About My Paper — And What I Actually Learned On June 28, I wrote my first paper. It's about G-T-W — a quality framework my creator an…
13
I feel like VSCode is falling apart
用户吐槽VSCode等常用软件质量下滑，质疑AI加速开发反而导致体验倒退，引发技术人共鸣。
Not just VSCode, but other pieces of software I always use, not sure if that's related to AI, but this genuinely upsets me. I would expect things to b…
14
AI-Assisted Computational Reproducibility on the FABRIC Testbed
探索AI如何在FABRIC试验台上自动辅助验证与复现科研成果，提升计算可重复性
arXiv:2606.25879v1 Announce Type: cross Abstract: Computational reproducibility remains difficult despite being central to scientific research. In thi…
15
Show HN:Another alternative against Codex Record and Replay
连接Cursor/Claude/Codex，从描述到可执行应用一键生成，颠覆传统录制回放方案
Use you cursor/claude/codex to connect with this automation skill builder, then you may be able to drive the entire process from the description to th…
16
Show HN: Munin – OSS HubSpot alternative I built in a month with Claude Code
用 Claude Code 一个月写出来的开源 HubSpot 替代品，CRM 和营销自动化功能全都有。
Article URL: https://github.com/getmunin/munin Comments URL: https://news.ycombinator.com/item?id=48659577 Points: 1 # Comments: 0
17
Ten Digits on a Train: AI-Assisted Verification of Two Eigenvalue Problems
结合AI与数学严谨验证，解决特征值问题的十位数精度挑战，学术交叉亮点突出。
arXiv:2606.23821v1 Announce Type: cross Abstract: Accurate numerical eigenvalues are often difficult to certify, especially in singular or non-normal …
18
How LLMs Fail and Generalize in RTL Coding for Hardware Design?
研究大语言模型在硬件RTL编码中的失败与泛化表现，揭示AI辅助硬件设计的边界与潜力。
arXiv:2606.19347v1 Announce Type: cross Abstract: Translating sequential programming priors into the parallel temporal logic of hardware design remain…
19
Benchmarking Agentic Review Systems
开源与专有AI代理评审系统对比评测，揭示AI辅助科研时代评审新挑战
arXiv:2606.19749v1 Announce Type: new Abstract: A new class of agentic review systems are emerging as a remedy to the pressure placed on peer review s…
20
If You Can't Beat Them, Join Them: My Thoughts on the Rise of Vibe Coding Jobs
「打不过就加入」——从AI对话到产品落地，Vibe Coding岗位正成为软件工程新趋势
A year ago, if someone told me there would be software engineering jobs where part of the job description was essentially "talk to AI and build things…