AI 资讯

Cloudflare details Town Lake, an internal unified data platform, and Skipper, an AI analytics agent unifying access to operational, billing, security, and business data. The platform processed ~91K billing queries, with billing forming majority usage. Built on a lakehouse architecture using Trino, Iceberg, R2, and DataHub, it enables governed cross-system analytics and natural language access. By Leela Kumili

At the event "The Briefing: AI for Science" earlier this week, Anthropic announced Claude Science, a new "AI workbench for scientists" that pulls fragmented tools and datasets into one environment, and generates figures and visuals. Anthropic, already dominating the industry with its popular coding tools and powerful AI models, framed the launch around what it […]

Get ready to navigate a free-for-all of social media handles.

YouTuber Jon Prosser has finally filed a formal response to Apple's lawsuit made against him and another defendant over allegedly stealing iOS secrets. In his response, Prosser denied that he "planned or participated in any conspiracy or coordinated scheme" for the "purpose of injuring Apple." However, Prosser admitted to recording a FaceTime call showing unreleased […]

A closer look at why users don’t need more tools in their daily lives. What they need are seamless integrations of useful features to match already existing, established mental models. Brought to you by Design Patterns For AI Interfaces , **friendly video course on UX** and design patterns by Vitaly.

In the AI slop-loaded, algorithm-powered modern reality, trends come and go - and the tech industry is no different. For the last few years, The Verge staff has compiled a selection of things that are IN for summer and OUT for summer - and each time there are some strong feelings. (Here are the last […]

A deep dive into what Anthropic, OpenAI, Perplexity and LangChain are actually building. Covering the orchestration loop, tools, memory, context management, and everything else that transforms a stateless LLM into a capable agent. You've built a chatbot. Maybe you've wired up a ReAct loop with a few tools. It works for demos. Then you try to build something production-grade, and the wheels come off: the model forgets what it did three steps ago, tool calls fail silently, and context windows fill up with garbage. The problem isn't your model. It's everything around your model. LangChain proved this when they changed only the infrastructure wrapping their LLM (same model, same weights) and jumped from outside the top 30 to rank 5 on TerminalBench 2.0. A separate research project hit a 76.4% pass rate by having an LLM optimize the infrastructure itself, surpassing hand-designed systems. That infrastructure has a name now: the agent harness. What Is the Agent Harness? The term was formalized in early 2026, but the concept existed long before. The harness is the complete software infrastructure wrapping an LLM: orchestration loop, tools, memory, context management, state persistence, error handling, and guardrails. Anthropic's Claude Code documentation puts it simply: the SDK is "the agent harness that powers Claude Code." OpenAI's Codex team uses the same framing, explicitly equating the terms "agent" and "harness" to refer to the non-model infrastructure that makes the LLM useful. The canonical formula, from LangChain's Vivek Trivedy: "If you're not the model, you're the harness." Here's the distinction that trips people up. The "agent" is the emergent behavior: the goal-directed, tool-using, self-correcting entity the user interacts with. The harness is the machinery producing that behavior. When someone says "I built an agent," they mean they built a harness and pointed it at a model. Beren Millidge made this analogy precise in his 2023 essay, Scaffolded LLMs as Natu

Built for the WeMakeDevs × Cognee "The Hangover Part AI" hackathon — Cognee Cloud track. ▶ Play it free: vegas-amnesia.vercel.app · ⭐ Code on GitHub The problem with most memory demos When you give a developer a memory API, the demo almost always looks the same: add() some documents, search() over them, print the answer. Two functions. It works, it's fine, and it teaches you almost nothing about why graph-based memory is different from stuffing everything into a context window. Cognee actually has a four-stage lifecycle — remember → recall → memify → forget — and the interesting parts are the two everyone skips. memify consolidates what you know into new inferences. forget lets you delete a belief and watch the graph heal around it. Memory you can reason over and correct . So instead of writing another RAG demo, I asked: what if the memory lifecycle wasn't the plumbing — what if it was the game ? Meet HAL-9001 You play HAL-9001 , a personal AI assistant (yes, HAL 9000's slightly more helpful successor). Your owner Dev had a wild night in Vegas. At 6 AM your memory graph was corrupted. His fiancée Priya lands at noon, there's a suspicious ring on his finger, and you remember nothing . The screen boots to a "MEMORY CORRUPTED" terminal and an empty graph. Your job: reconstruct the night, catch the lies, and answer the final question — what happened, and where's the ring? — before noon. Every location you explore, every clue you examine, every witness you interrogate feeds a live 3D memory graph that you can pop open at any time. That graph isn't a visualization of the game state. It is the game state — it's your Cognee dataset, rendered. The four mechanics = the four lifecycle ops Here's the mapping I'm most proud of. Each Cognee operation is a verb the player performs: You do this in-game Cognee Cloud call What happens 🗂 File It on a clue POST /api/v1/remember The fact is ingested + auto-cognified into graph nodes that pop into view ❓ Ask HAL a question POST /api/v1/r

Here are five architectural lessons we've learned from designing software for modern retailers.* Designing ERP Software for Retail: Five Lessons Every Software Engineer Should Know When people hear the word ERP , they often think of accounting software, dashboards, or inventory management. As software engineers, we see something different. We see distributed systems. Complex business workflows. Real-time data synchronization. Concurrent transactions. Event-driven architecture. And perhaps the biggest challenge of all—representing how real businesses actually operate. At RetailWings , we've learned that building an ERP for retail isn't simply a software engineering challenge. It's a business engineering challenge. Here are five lessons every engineer should understand before designing an ERP platform for modern retail. 1. Retail Doesn't Run in Modules—It Runs as One Business One of the biggest architectural mistakes in business software is treating departments as isolated applications. Many systems separate: Sales Inventory Finance Procurement HR But retailers don't experience their businesses that way. One sale immediately affects inventory. Inventory influences procurement. Procurement impacts finance. Finance drives reporting. Everything is connected. A well-designed ERP should reflect these relationships rather than forcing departments into disconnected silos. 2. Inventory Is More Than a Database Table To many engineers, inventory may appear to be a simple CRUD problem. Create. Read. Update. Delete. Retail quickly proves otherwise. Inventory changes through: Sales Returns Transfers Damages Procurement Stock adjustments Warehouse movements Manual reconciliations Every movement has financial implications. Every movement must be traceable. Designing inventory requires thinking in terms of events, not just records. 3. Real-Time Data Changes Everything Retail managers don't want yesterday's reports. They want answers now. How much stock is left? Which branch is sellin

I have been turning old code projects into sellable source-code products. The hard part is not changing the cover image. It is not renaming the ZIP. It is deciding whether the project deserves to be sold at all. A lot of old apps are useful to the person who built them. Far fewer are useful to a stranger who has never seen the repo, never heard the backstory, and only wants to know one thing: Will this save me time, or will it become another folder I regret buying? Here is the checklist I now use before treating a source-code project as a starter kit. 1. The buyer must understand the workflow "Full-stack dashboard" is not enough. A buyer should immediately understand the workflow the project helps with. For example: a review and scoring portal; a maintenance and work-order dashboard; an email-template governance tool; a runnable technical code lab. The more generic the product sounds, the harder it is to buy. I now try to answer this in one sentence: This kit helps [specific buyer] start from a working foundation for [specific workflow]. If I cannot fill that sentence honestly, the project is not ready. 2. A stranger must be able to run it "It runs on my machine" is not a product standard. A buyer needs a path from download to working state. That usually means: setup instructions; environment notes; seeded demo data; demo accounts or fixtures; expected local startup behavior; known limitations; a simple smoke-test checklist. The goal is not perfection. The goal is that a competent developer should not have to reverse-engineer the project before deciding whether it is useful. 3. The product needs proof, not adjectives Marketing adjectives are cheap: production-ready; powerful; scalable; enterprise-grade; battle-tested. Most of those words create more risk than trust if they are not backed by evidence. Better proof looks boring: screenshots; a short demo video; a verified release ZIP; install notes; architecture notes; included / not-included boundaries; a changelog;

We assume you already know how to write simple Python programs and understand basic syntax (if, for, functions, lists). Here, we're not discussing how to use the language, but why it works the way it does. In the previous article, we learned that variables in Python don't store data themselves. Instead, a name simply refers to an object. With numbers, this feels quite intuitive: x = 10 y = x Both names refer to the same object whose value is 10 . So far, everything seems straightforward. But as soon as we start working with lists, Python's behavior often surprises people. Consider this example: a = [ 1 , 2 ] b = a a . append ( 3 ) print ( b ) Many beginners expect the output to be: [ 1 , 2 ] After all, we modified a , not b . Instead, Python prints: [ 1 , 2 , 3 ] It looks as if changing one variable somehow changed the other. At first glance, it may seem like Python is keeping the two variables synchronized behind the scenes. In reality, the explanation is much simpler. Let's go back to the idea from the previous article. When we write: a = [ 1 , 2 ] Python creates a list object, and the name a becomes bound to it. Visually: a ───► [1, 2] Next comes this line: b = a This is the crucial part. Python does not create a second list. No copy is made. Instead, the name b is bound to exactly the same object. Now the picture looks like this: a ─┐ ├──► [1, 2] b ─┘ Notice that there's still only one list. The only thing that changed is that there are now two names referring to it. That's why the next line: a . append ( 3 ) doesn't create a new list. It modifies the existing object. After the call to append() , the picture becomes: a ─┐ ├──► [1, 2, 3] b ─┘ Since both names refer to the same object, the change is visible through both of them. In other words, Python didn't modify two different lists. It modified one list that simply has two names. This behavior is one of the most common sources of confusion for beginners. For example, you might write a function that modifies a l

Table of contents What is OAUTH A trip to OAUTH1.0Ville What is OAUTH2.0 Examples of OAUTH Technology OIDC Hands-on Implementation with Microsoft Entra ID What is OAUTH OAUTH is a technological standard that allows you to authorize one app or service to sign in to another without divulging private information, such as passwords. OAUTH stands for Open-Authorization , not Authentication . Authentication is a process that verifies your identity, although OAUTH involves identity verification, its main purpose is to grant access to connect you with different apps and services without requiring you to create a new account. How Does OAUTH Work OAUTH uses access tokens, and this is what makes OAUTH secure to use. An access token is a piece of data that contains information about the user and the resource the token is intended for. A token will also include specific rules for data sharing . For example, you want to share your photos from Instagram with Kyrier — An intelligent email platform built for professionals who refuse to let their inbox run their day , but you only want Kyrier to access your profile image. Kyrier does not also need to access your direct messages or friends list. Instagram issues an access token to Kyrier to access the data you approve (your profile image in this case) on your behalf. So an access token will only allow Kyrier to access your profile image, not even other photos on your page. There may be rules governing when Kyrier can use the access token, it might be for a single use or for recurring uses, and it always has an expiration date. A trip to OAUTH1.0Ville Welcome to OAuth1.0Ville. Please keep your hands inside the vehicle. This is where OAuth started. It was built only for websites , back when "an app" meant a web page and nothing else. Although it worked, it had a lot of problems: Only three authorization flows (2.0 has six) No real plan for mobile or modern apps A scaling problem it never solved It also makes you cryptographically sign e

Welcome to My Generative AI Learning Journey Artificial Intelligence is changing the way we work, learn, build software, and solve problems. Every day, new AI tools, models, and technologies are being released, making it difficult to know where to begin. Instead of randomly watching videos or reading articles, I've decided to follow a structured learning path—and I'm inviting you to join me. This blog marks the beginning of a long-term Generative AI learning series. Whether you're a student, software developer, freelancer, entrepreneur, or simply curious about AI, this roadmap will help you understand what we'll learn together over the coming weeks and months. The goal isn't just to understand AI theory. It's to build practical skills that can be used in real-world projects and professional development. Why Learn Generative AI in 2026? Generative AI is no longer a futuristic concept. It is already transforming industries such as: Software Development Healthcare Education Finance Marketing Customer Support E-commerce Human Resources Design and Creativity Companies are actively seeking professionals who can build AI-powered applications, automate workflows, and integrate AI into existing systems. Learning Generative AI today means preparing for the next generation of technology. What You Can Expect from This Series This series is designed for beginners but will gradually move toward advanced concepts. Each article will build upon the previous one, making the learning process simple and structured. We'll focus on: Understanding AI concepts Learning industry terminology Exploring popular AI models Writing effective prompts Building AI applications Working with APIs Using open-source models Creating AI-powered software Deploying AI projects By the end of this journey, you'll have both theoretical knowledge and practical development experience. Complete Learning Roadmap Phase 1: AI Fundamentals We'll begin by building a strong foundation. Topics include: What is Generativ

Prompt analytics and citation mapping for AI search Discussion | Link

企业尽调智能体实战：60+真实企业的AI尽调报告 从5天到10分钟：AI如何重构企业尽调 企业贷前尽调，银行和金融机构最头疼的环节。一位信贷经理曾这样描述他的工作：打开天眼查查工商信息，切到Wind拉行情，再打开百度搜新闻，最后把散落在七八个系统里的数据拼进Word模板。一家企业，至少5天。如果碰上集团客户、关联方众多的，两周起步。 一家支行行长曾无奈地说："25个客户经理，每个人做的尽调报告格式都不一样。同样的企业，A经理评'低风险'，B经理评'中等风险'，谁对谁错无从判断。"问题的根源不是人的能力差异，而是工具链的碎片化——数据散落在不同系统里，没有统一入口，也没有标准化的采集流程。 我们调研了12家金融机构的尽调流程，发现三个共性痛点： 信息散落 （数据分布在6-10个系统中）、 耗时漫长 （单家企业5-10个工作日）、 质量参差 （依赖个人经验，无标准化流程）。 本文记录的，是一个用AI Agent解决这个问题的实战项目——企业尽调引擎v5.0。它不是概念验证，不是Demo，而是在60+家真实企业上跑通的生产级系统。 技术架构：多源数据整合的数据流 尽调的核心难题不是"分析"，而是"采集"。一家上市公司的完整画像，需要从至少6个异构数据源拉取信息。传统方式是人肉Copy-Paste，我们的方案是用Agent自动编排数据流： 用户输入 "美的集团" │ ▼ ┌─────────────────────────────────┐ │ Step 1: 股票代码查询 │ │ 联网搜索 → 000333.SZ │ └──────────────┬──────────────────┘ │ ┌──────────┴──────────┐ ▼ ▼ ┌─────────┐ ┌──────────┐ │ Step 2a │ │ Step 2b │ │ 实时行情 │ │ 新闻舆情 │ │ ifind │ │ 联网搜索 │ └────┬────┘ └─────┬────┘ │ │ └─────────┬──────────┘ │ ┌──────────┼──────────┐ ▼ ▼ ▼ ┌────────┐ ┌────────┐ ┌────────┐ │Step 3a │ │Step 3b │ │Step 3c │ │工商信息 │ │风险扫描 │ │估值指标 │ │ MCP │ │ MCP │ │ MCP │ └───┬────┘ └───┬────┘ └───┬────┘ │ │ │ └──────────┼──────────┘ │ ▼ ┌─────────────────────────────────┐ │ Step 4: 舆情分析 + 综合评分 │ │ 多源交叉验证 → 生成尽调报告 │ │ 输出: JSON(5KB) + Markdown(4KB) │ └─────────────────────────────────┘ 这个数据流的核心设计原则是 并行采集、串行推理 。Step 2的行情和舆情可以并行获取，Step 3的三个MCP调用也可以并行，但Step 4的综合评分必须等所有数据到齐后才能做交叉验证。这种设计把端到端耗时压到了10分钟以内。 另一个关键设计是 渐进式降级 ：如果MCP工具不可用（比如企业是非上市公司），引擎会跳过行情和估值模块，仅返回工商+风险+新闻的"基础版"报告，而不是直接报错退出。这一设计在实际使用中至关重要——我们的60+企业样本中，有11家是非上市企业，如果要求所有数据源齐备才能出报告，这11家就会被拒之门外。 五大能力详解 1. 股票代码查询 输入企业名称，自动搜索匹配股票代码。比如输入"美的集团"，引擎通过联网搜索拿到 000333.SZ 。这个步骤看似简单，却是后续所有数据获取的前提——行情、估值、历史走势全部依赖股票代码。对于非上市企业，引擎会标记 stock_code: null 并跳过相关模块。在实际测试中，股票代码查询的成功率超过98%，少数失败案例主要是名称变更（如"格力地产"更名为"珠免集团"）尚未被搜索引擎索引。 2. 实时行情数据 通过ifind接口获取实时股价、涨跌幅、成交量、换手率等指标。这些数据直接写入报告的"行情数据"章节，避免分析师手动从交易软件抄录。更重要的是，行情数据与后续的估值指标做交叉验证——如果PE_TTM显示14倍但股价异常波动，报告会标注"数据一致性待确认"。 3. 企业新闻舆情 联网搜索获取企业最新新闻，引擎对新闻做情感分析后输出舆情等级（正面/中性/负面）和舆情得分（0-100）。这不是简单的关键词匹配，而是基于上下文的语义判断。当正面信号和风险信号同时出现时，报告会分别列出，而非简单抵消。一条"美的集团海外营收创新高"和一条"美的集团遭反倾销

4A企业架构+TOGAF如何指导Agent Skill设计 引言：AI Skill设计的"巴别塔"困局 当下的AI Agent生态，正陷入一种似曾相识的混乱。 去年帮一家保险公司梳理Agent技能库，发现100多个Skill横七竖八地堆在一起——有的直接调API，有的内嵌业务逻辑，有的把数据获取和分析揉成一团。问架构师这些Skill怎么分类，回答是"按安装顺序排的"。再问两个Skill之间数据怎么流转，回答是"各写各的"。一个股票监控Skill自己爬数据、自己做分析、自己发消息，三件事耦合在同一个脚本里。换一个场景想复用其中的分析逻辑？做不到，只能重写。 这不是个例。几乎所有率先部署AI Agent的企业都面临同样的困境：Skill越堆越多，越堆越乱。缺乏统一的能力域划分，缺乏标准化的数据接口，缺乏清晰的组合规则，缺乏可复用的构建块沉淀。 听起来很熟悉？没错——这正是企业架构在20年前要解决的问题。当年企业信息化的混乱，和今天AI Skill的混乱，本质上是一回事：没有架构约束的开发，必然走向无序。 4A企业架构（业务架构BA、数据架构DA、应用架构AA、技术架构TA）加上TOGAF的构建块思想，为Agent Skill设计提供了一套经过验证的方法论。本文试图建立这二者之间的映射框架，并用实际案例说明其可行性。 4A映射框架：四个问题驱动Skill设计 企业架构的核心是四个问题：做什么（BA）、数据怎么流（DA）、用什么组合（AA）、底层怎么支撑（TA）。这四个问题同样适用于Skill设计。 ┌───────────────────────────────────────────────────────────────┐ │ 4A → Skill 映射框架 │ ├───────────────────────────────────────────────────────────────┤ │ │ │ BA 业务架构 │ │ ┌─────────────────────────────────────────────────────────┐ │ │ │ Skill的业务能力域划分、价值链映射 │ │ │ │ → 回答"这套Skill体系解决什么业务问题" │ │ │ └────────────────────────────┬────────────────────────────┘ │ │ │ │ │ DA 数据架构 │ │ │ ┌────────────────────────────▼────────────────────────────┐ │ │ │ Skill的数据流、信息交换标准 │ │ │ │ → 回答"Skill之间数据怎么流转、用什么格式" │ │ │ └────────────────────────────┬────────────────────────────┘ │ │ │ │ │ AA 应用架构 │ │ │ ┌────────────────────────────▼────────────────────────────┐ │ │ │ Skill的组合关系、依赖图谱 │ │ │ │ → 回答"哪些Skill可以组装、依赖关系是什么" │ │ │ └────────────────────────────┬────────────────────────────┘ │ │ │ │ │ TA 技术架构 │ │ │ ┌────────────────────────────▼────────────────────────────┐ │ │ │ Skill的运行时、工具链、基础设施 │ │ │ │ → 回答"Skill跑在什么环境上、需要哪些依赖" │ │ │ └─────────────────────────────────────────────────────────┘ │ │ │ └───────────────────────────────────────────────────────────────┘ 用表格更清晰地展示每一层的核心映射关系： 架构层 核心问题 Skill映射 示例 BA 业务架构 Skill解决什么业务问题？ 按能力域分组：协同办公、内容生产、数据智能、系统运维等 "数据智能"域包含客户画像、股票监控、财务智能等Skill DA 数据架构 Skill间数据如何流转？ 统一数据格式(Markdown/JSON)，标准化数据源→转换→消费管道 搜索Skill输出Markdown，报告Skill消费Markdown生成PDF AA 应用架构 Skill如何组合复用？ 构建依赖图谱，上层组合下层，同类可替换 日报Skill = 搜索 + 摘要 + PDF转换 + 消息推送