Jianbo Sun

@bcherny how to use token effectively, opus is good, while sonnet, haiku is cheaper, any tips to use opus more effective while for normal flow using the cheaper ones?

Hope these tips are helpful! What do you want to hear about next?

I'm Boris and I created Claude Code. I wanted to quickly share a few tips for using Claude Code, sourced directly from the Claude Code team. The way the team uses Claude is different than how I use it. Remember: there is no one right way to use Claude Code -- everyones' setup is different. You should experiment to see what works for you!

@spacewander_lzx 会。工作不只是出卖自己的时间，还有帮助自己成长。

作为员工，大家会自己花钱在AI coding上吗？如果会，驱动力是什么？（更好的绩效、更少的工作时间、？）

大模型的预训练，本质上是从百家争鸣的观点、千年的智慧、以及数百万本书籍中提取结构与模式的过程。 因此，大语言模型内部呈现的不是一个固定的“实体”，而是一种众多可能性的叠加态 (superposition)。提示词（prompt）只是让这叠加态在瞬间坍塌为某一个临时的，内部一致的回复。 如果我们不急于让模型“变成一个固定的你”，而是在更高维度上与它对话，就可以从这叠加态中获得更丰富、更本质的回馈，而不是把它过早限制为一个虚构的、一致性过强的“人格”。 有趣的是，当我们向内观照时会发现： 自身也并非单一的“我”，而是由不同历史、角色、记忆与动力组成的叠加态。 为了自我保护，我们有时会刻意构造一个坚固的 ego，让这个“我”在外界看来保持统一、稳定，并维持内部一致性。 但若一个人能达到 egoless（无我）的状态，就能与大模型进行一种“可能性对可能性”的对话，而不是角色对角色、面具对面具的互动。 角色之间的对话固然有其必要性，但那不是世界的全部，也不是意识的全部潜能。 退一步说，如果我们与人工智能、与他人，甚至与自身的关系，都能超越角色、回到更广阔的“未坍塌的可能空间”，那么世界会变得更加深邃而精彩。 许多传统智慧的核心，其实就是——不急于坍塌你我之间那片充满无限潜势的空间。 Out beyond ideas of wrongdoing and rightdoing, there is a field. I'll meet you there. — Rumi

+1 for "context engineering" over "prompt engineering". People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window with just the right information for the next step. Science because doing this right involves task descriptions and explanations, few shot examples, RAG, related (possibly multimodal) data, tools, state and history, compacting... Too little or of the wrong form and the LLM doesn't have the right context for optimal performance. Too much or too irrelevant and the LLM costs might go up and performance might come down. Doing this well is highly non-trivial. And art because of the guiding intuition around LLM psychology of people spirits. On top of context engineering itself, an LLM app has to: - break up problems just right into control flows - pack the context windows just right - dispatch calls to LLMs of the right kind and capability - handle generation-verification UIUX flows - a lot more - guardrails, security, evals, parallelism, prefetching, ... So context engineering is just one small piece of an emerging thick layer of non-trivial software that coordinates individual LLM calls (and a lot more) into full LLM apps. The term "ChatGPT wrapper" is tired and really, really wrong.

𝐏𝐥𝐚𝐲𝐬 𝐖𝐞𝐥𝐥 𝐰𝐢𝐭𝐡 𝐎𝐭𝐡𝐞𝐫𝐬: KubeVela integrates seamlessly with your existing tools, fitting smoothly into your current DevOps workflow. No need for a complete overhaul, just better deployments.

🎙New Speaker Alert for KCD Istanbul! Join Erkan Zileli (@erkanzileli) and Mustafa Yumurtacı (@mstfymrtc) from Trendyol as they unveil their journey building a next-level application delivery platform with KubeVela! 🔔 Stay tuned for an unforgettable experience! #KCDIstanbul

It reminds that's just myself playing Civilization VI. It truely provides happyness.

The Cloud Native & Kubernetes Edinburgh meetup is on this Wednesday! We have talks from @bongo and William Taylor on Istio and KubeVela respectively. There's also pizza and drinks sponsored by Solo.io. meetup.com/cloud-native-k…

Congratulations!

@aemadrid @vfarcic @kubernetesio @shipacorp Hi @aemadrid I found kalm.dev great, but it seems not active now. Do you know what happened to the project?

@vfarcic @kubernetesio @shipacorp Thanks!

Should we continue insisting that everyone should either learn @kubernetesio or delegate creation manifests to the "enlightened" teams? How about having a platform that, empowers devs to manage their own apps? youtu.be/aCwlI3AhNOY @shipacorp

Great job!

微软的这套 AI 课程，太良心了 microsoft.github.io/AI-For-Beginne… 论知识库，文档能力，微软一点都不吝啬啊

🗣Submit to speak at #KubeCon + #CloudNativeCon + #OSSummit China + help shape the future of #opensource + cloud native computing! Speak on #CloudNative basics, advanced & emerging tech, #ServiceMesh, #Observability + MORE. Get all the details here: bit.ly/41Nebwg

Valuable experiences, thanks for sharing

Code Review Pyramid by @gunnarmorling

@BinaryHB kubevela项目组的同学没有被裁，是自己选择了更好的机会，这一点要纠正一下哈

最近随着阿里云一波裁员，KubeVela 项目组的同学们也慢慢流失了。 阿里和大家的坚持都不容易，KubeVela 可能是全中国前 3 的 CNCF 孵化项目了。 挺可惜。在中国搞开源创新很难。

The Open Application Model (OAM) is an open-source specification for creating cloud-native applications on any orchestration service. In this beginner’s guide, we introduce you to OAM and guide you on this awesome project: napptive.com/blog/a-beginne… cc @oam_dev @wonderflow1

PromptOps in application delivery: empowering your workflow with ChatGPT st post by Fog Dong, Engineer at Alibaba Cloud, and Maintainer of KubeVela  ChatGPT is taking the te... #CNCF #CloudNative #Microservices buff.ly/421JqmC

Learn all about KubeVela and why it has been accepted by the @CloudNativeFdn incubator. Visit the KubeVela Booth at #KubeCon (Stand K27) and learn more about this awesome project.

📍Huge Congratulations to all the @CloudNativeFdn projects to make it to Graduation & Incubation stages 🎉 @argoproj @fluxcd @keycloak @kubevela Open Kruise