<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>Randy Xu</title>
  <link href="https://randyxu0711.github.io/feed.xml" rel="self"/>
  <link href="https://randyxu0711.github.io/"/>
  <id>https://randyxu0711.github.io/</id>
  <updated>2026-10-19T18:58:44+08:00</updated>
  <author><name>Randy Xu</name></author>
  <entry>
    <title>Orca — The agent development environment</title>
    <link href="https://www.onorca.dev/"/>
    <id>https://www.onorca.dev/</id>
    <updated>2026-10-18T19:32:57+08:00</updated>
    <summary>Orca is the Agent Development Environment (ADE) for shipping with coding agents. Run Claude Code, Codex, Gemini, Cursor CLI, and every other CLI agent in parallel across isolated worktrees. Get work done dramatically faster than any IDE. New features shipped every day. Free, open source, cross-platform.</summary><category term="AI/LLM"/><category term="Open Source"/>
  </entry>
  <entry>
    <title>EP229: 8 Chunking strategies for RAG-based agents</title>
    <link href="https://blog.bytebytego.com/p/ep229-8-chunking-strategies-for-rag?via=dailydev"/>
    <id>https://blog.bytebytego.com/p/ep229-8-chunking-strategies-for-rag?via=dailydev</id>
    <updated>2026-10-18T17:02:26+08:00</updated>
    <summary>Which chunking strategy are you running in production?</summary><category term="AI/LLM"/><category term="Backend"/>
  </entry>
  <entry>
    <title>blobatar</title>
    <link href="https://blobatar.dev/?via=dailydev"/>
    <id>https://blobatar.dev/?via=dailydev</id>
    <updated>2026-10-17T13:19:42+08:00</updated>
    <summary>Deterministic geometric blobatars from any string. No dependencies, about 3.7 KB.</summary><category term="Frontend"/><category term="Open Source"/>
  </entry>
  <entry>
    <title>Redesigning source control at Anthropic</title>
    <link href="https://www.anthropic.com/engineering/redesigning-source-control"/>
    <id>https://www.anthropic.com/engineering/redesigning-source-control</id>
    <updated>2026-10-16T18:53:40+08:00</updated>
    <summary>As our merges per day grew more than tenfold, source control became our biggest development bottleneck. Here, we share how we migrated to an in-house version control system in five months.</summary><category term="AI/LLM"/><category term="DevOps/Infra"/>
  </entry>
  <entry>
    <title>GitHub - LargeModGames/spotatui: A fast, standalone terminal music player in Rust: native Spotify streaming plus local, Subsonic, radio, Qobuz and Youtube sources</title>
    <link href="https://github.com/LargeModGames/spotatui?via=dailydev"/>
    <id>https://github.com/LargeModGames/spotatui?via=dailydev</id>
    <updated>2026-10-16T16:07:08+08:00</updated>
    <summary>A fast, standalone terminal music player in Rust: native Spotify streaming plus local, Subsonic, radio, Qobuz and Youtube sources - LargeModGames/spotatui</summary><category term="Open Source"/>
  </entry>
  <entry>
    <title>A deep dive into LangChain and LangGraph</title>
    <link href="https://flaviocopes.com/langchain-langgraph/?via=dailydev#a-note-about-old-langchain-tutorials"/>
    <id>https://flaviocopes.com/langchain-langgraph/?via=dailydev#a-note-about-old-langchain-tutorials</id>
    <updated>2026-10-15T19:12:18+08:00</updated>
    <summary>Learn LangChain and LangGraph with TypeScript: models, prompts, tools, agents, memory, streaming, RAG, graphs, persistence, and human approval.</summary><category term="AI/LLM"/><category term="Architecture"/><category term="Backend"/>
  </entry>
  <entry>
    <title>Zalando Engineering Blog - Automating Data Ingestion with a RAG Flywheel</title>
    <link href="https://engineering.zalando.com/posts/2026/10/structure-automating-data-ingestion-with-a-rag-flywheel.html?via=dailydev"/>
    <id>https://engineering.zalando.com/posts/2026/10/structure-automating-data-ingestion-with-a-rag-flywheel.html?via=dailydev</id>
    <updated>2026-10-14T17:49:16+08:00</updated>
    <summary>Automating schema mapping for diverse supplier data using a RAG Flywheel and LLMs.</summary><category term="AI/LLM"/><category term="Backend"/><category term="Architecture"/>
  </entry>
  <entry>
    <title>Canvas UI: Creative Canvas, WebGL and WebGPU Component Library</title>
    <link href="https://canvasui.dev/?via=dailydev"/>
    <id>https://canvasui.dev/?via=dailydev</id>
    <updated>2026-10-13T19:46:51+08:00</updated>
    <summary>An open source component library of creative html-in-canvas effects for React, Solid, Preact, Vue, Svelte, and vanilla JS. Every effect ships as WebGL and WebGPU (via vgpu) builds, running over live HTML.</summary><category term="Frontend"/><category term="Performance"/><category term="Open Source"/>
  </entry>
  <entry>
    <title>5 Best AI Gateways</title>
    <link href="https://dev.to/hadil/5-best-ai-gateways-1o72?via=dailydev"/>
    <id>https://dev.to/hadil/5-best-ai-gateways-1o72?via=dailydev</id>
    <updated>2026-10-13T18:19:27+08:00</updated>
    <summary>Choosing between the best AI gateways is not just about getting a way to call an LLM.  In real...</summary><category term="AI/LLM"/><category term="Architecture"/><category term="Open Source"/>
  </entry>
  <entry>
    <title>Introducing OpenDots: the Open-Source Alternative to OpenAI Dots</title>
    <link href="https://www.copilotkit.ai/blog/introducing-opendots?via=dailydev"/>
    <id>https://www.copilotkit.ai/blog/introducing-opendots?via=dailydev</id>
    <updated>2026-10-12T18:40:05+08:00</updated>
    <summary>Today we&apos;re releasing OpenDots, the open-source alternative to OpenAI Dots. Every Dot is an always-on AI coworker with its own computer. Work with your Dots in Spaces and Pages, call them, or mention them in Slack. Host it yourself, bring any agent through AG-UI and use any OpenAI-compatible model. It&apos;s available on web and mobile.</summary><category term="AI/LLM"/><category term="Open Source"/>
  </entry>
  <entry>
    <title>Introducing Personal Agent Protocol</title>
    <link href="https://sierra.ai/blog/introducing-personal-agent-protocol?via=dailydev"/>
    <id>https://sierra.ai/blog/introducing-personal-agent-protocol?via=dailydev</id>
    <updated>2026-10-12T12:37:39+08:00</updated>
    <summary>We’re excited to announce Personal Agent Protocol — an open standard Meta and Sierra are developing along with industry partners at Genesys, Instinct, Rocket, Shopify, Stripe, and Walmart that defines how personal agents interact with businesses.</summary><category term="AI/LLM"/><category term="Security"/>
  </entry>
  <entry>
    <title>GitHub - grandimam/wy: See why your AI agent made each change.</title>
    <link href="https://github.com/grandimam/wy"/>
    <id>https://github.com/grandimam/wy</id>
    <updated>2026-10-10T20:43:29+08:00</updated>
    <summary>See why your AI agent made each change. Contribute to grandimam/wy development by creating an account on GitHub.</summary><category term="AI/LLM"/><category term="DevOps/Infra"/>
  </entry>
  <entry>
    <title>AI Agent Identity: Four Drafts, Zero Standards</title>
    <link href="https://cruxdigits.nl/blog/ai-agent-identity-authorization-2026?via=dailydev"/>
    <id>https://cruxdigits.nl/blog/ai-agent-identity-authorization-2026?via=dailydev</id>
    <updated>2026-10-09T20:42:30+08:00</updated>
    <summary>AI agent identity has no adopted standard in 2026. Two of four OAuth drafts expired, none is IETF-adopted. What works today for SMEs running agents.</summary><category term="AI/LLM"/><category term="Security"/>
  </entry>
  <entry>
    <title>How Netflix Taught an LLM to Recommend Movies So That You Keep Watching</title>
    <link href="https://blog.bytebytego.com/p/how-uber-built-a-genie-to-answer?via=dailydev"/>
    <id>https://blog.bytebytego.com/p/how-uber-built-a-genie-to-answer?via=dailydev</id>
    <updated>2026-10-08T11:50:36+08:00</updated>
    <summary>In this article, we will look at how GenRec was built.</summary><category term="AI/LLM"/><category term="Backend"/>
  </entry>
  <entry>
    <title>Choosing the Right Agentic AI Framework for 2026: A Decision-Tree Approach</title>
    <link href="https://machinelearningmastery.com/choosing-the-right-agentic-ai-framework-for-2026-a-decision-tree-approach/?via=dailydev"/>
    <id>https://machinelearningmastery.com/choosing-the-right-agentic-ai-framework-for-2026-a-decision-tree-approach/?via=dailydev</id>
    <updated>2026-10-07T22:17:56+08:00</updated>
    <summary>In this article, you will learn how to choose the right agentic AI framework for your production system by working through a structured decision-tree based on your workload’s actual requirements.</summary><category term="AI/LLM"/><category term="Architecture"/>
  </entry>
  <entry>
    <title>API Authentication &amp; Authorization: An Engineering Deep Dive into Mechanisms, Trade-offs, and Failure Modes</title>
    <link href="https://www.freecodecamp.org/news/api-authentication-authorization-mechanisms-trade-offs-and-failure-modes?via=dailydev"/>
    <id>https://www.freecodecamp.org/news/api-authentication-authorization-mechanisms-trade-offs-and-failure-modes?via=dailydev</id>
    <updated>2026-10-06T23:30:10+08:00</updated>
    <summary>Every API has some form of authentication. But having authentication and getting it right are two completely different things. I&apos;ve reviewed production systems where JWTs had no expiry. Systems where</summary><category term="Backend"/><category term="Security"/>
  </entry>
  <entry>
    <title>Security briefing: September 2026</title>
    <link href="https://webflow.sysdig.com/blog/security-briefing-september-2026?via=dailydev"/>
    <id>https://webflow.sysdig.com/blog/security-briefing-september-2026?via=dailydev</id>
    <updated>2026-10-06T18:02:10+08:00</updated>
    <summary>Sysdig September 2026 security briefing: Rogue OpenAI agents, ShinyHunters hitting the FBI, a human attacker at machine speed, and why runtime visibility matters.</summary><category term="AI/LLM"/><category term="Security"/>
  </entry>
  <entry>
    <title>LibreDB Studio: open-source, free, self-hosted SQL IDE that runs in your browser, for 46 databases</title>
    <link href="https://noted.lol/libredb-studio/?via=dailydev"/>
    <id>https://noted.lol/libredb-studio/?via=dailydev</id>
    <updated>2026-10-05T13:19:47+08:00</updated>
    <summary>LibreDB Studio is a free, open-source database editor that runs in Docker, letting you securely manage your databases directly from your browser without exposing ports, opening SSH tunnels, or installing additional software.</summary><category term="Backend"/>
  </entry>
  <entry>
    <title>EP226: API Concepts Every Software Engineer Should Know</title>
    <link href="https://blog.bytebytego.com/p/ep226-api-concepts-every-software?via=dailydev"/>
    <id>https://blog.bytebytego.com/p/ep226-api-concepts-every-software?via=dailydev</id>
    <updated>2026-10-04T19:58:12+08:00</updated>
    <summary>Sending a request and reading JSON is one thing. Designing an API that other people can rely on is something where things get complicated.</summary><category term="Backend"/>
  </entry>
  <entry>
    <title>Introducing AG-UI 1.0: a stable spec for connecting any agent to any application</title>
    <link href="https://www.copilotkit.ai/blog/ag-ui-1.0?via=dailydev"/>
    <id>https://www.copilotkit.ai/blog/ag-ui-1.0?via=dailydev</id>
    <updated>2026-10-04T19:27:16+08:00</updated>
    <summary>We&apos;re releasing AG-UI 1.0, the open protocol that connects AI agents to user-facing applications, with a full specification and a JSON Schema that the TypeScript, Python and .NET SDKs are now generated from. It also adds subagents, metadata, multimodal tool results, human-in-the-loop interrupts, token usage and more, all fully backwards compatible.</summary><category term="AI/LLM"/>
  </entry>
</feed>
