<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Bil Arikan</title><description>Posts and notes from Bil Arikan on learning, AI, and getting systems, people, and data to work together.</description><link>https://bil.arikan.ca/</link><item><title>Case study : from scattered documents to a pilot</title><link>https://bil.arikan.ca/posts/agent-ready-knowledge-case-study-scattered-documents-to-a-pilot/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/agent-ready-knowledge-case-study-scattered-documents-to-a-pilot/</guid><description>What an applied-AI team can build for a sales organization : a prototype enablement agent, what went wrong along the way, and a starter kit for the next team.</description><pubDate>Mon, 05 Oct 2026 04:00:00 GMT</pubDate><category>case-study</category><category>ai-agents</category><category>sales-enablement</category><category>knowledge-management</category><category>learning-design</category><category>Agent-Ready Knowledge</category></item><item><title>Putting an agent next to a sales rep --- agent architecture, data model, security/privacy, and cost structure</title><link>https://bil.arikan.ca/posts/agent-ready-knowledge-putting-an-agent-next-to-a-sales-rep/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/agent-ready-knowledge-putting-an-agent-next-to-a-sales-rep/</guid><description>The harness, data model, security and privacy points, and cost of a real-time sales assistant built with Google ADK, and the break-even it has to clear.</description><pubDate>Wed, 30 Sep 2026 04:00:00 GMT</pubDate><category>ai-agents</category><category>google-adk</category><category>architecture</category><category>security</category><category>economics</category><category>sales-enablement</category><category>Agent-Ready Knowledge</category></item><item><title>LLMs reward expertise, not prompting tricks</title><link>https://bil.arikan.ca/notes/llms-reward-expertise/</link><guid isPermaLink="true">https://bil.arikan.ca/notes/llms-reward-expertise/</guid><description>The ceiling on what you get out of an LLM is set by your own domain expertise, not your prompting technique.</description><pubDate>Tue, 29 Sep 2026 13:00:00 GMT</pubDate><category>ai</category><category>llms</category><category>expertise</category></item><item><title>Eight fallacies of distributed computing, 21 years on</title><link>https://bil.arikan.ca/notes/fallacies-of-distributed-computing/</link><guid isPermaLink="true">https://bil.arikan.ca/notes/fallacies-of-distributed-computing/</guid><description>Two decades on, the eight fallacies of distributed computing still show up as buried assumptions in how systems get designed.</description><pubDate>Tue, 29 Sep 2026 12:00:00 GMT</pubDate><category>distributed-systems</category><category>networking</category></item><item><title>Making music with code: Strudel</title><link>https://bil.arikan.ca/notes/making-music-with-code-strudel/</link><guid isPermaLink="true">https://bil.arikan.ca/notes/making-music-with-code-strudel/</guid><description>Strudel is a JavaScript port of Tidal Cycles for live-coding music right in the browser.</description><pubDate>Tue, 29 Sep 2026 11:00:00 GMT</pubDate><category>music</category><category>live-coding</category></item><item><title>Turning narrated demos into scalable agentic knowledge</title><link>https://bil.arikan.ca/posts/agent-ready-knowledge-narrated-demos-into-agentic-knowledge/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/agent-ready-knowledge-narrated-demos-into-agentic-knowledge/</guid><description>Instead of asking a team to watch recordings they will forget tomorrow, I turned two narrated demos into knowledge an agent serves at the moment of need.</description><pubDate>Sat, 26 Sep 2026 04:00:00 GMT</pubDate><category>ai-agents</category><category>knowledge-capture</category><category>sales-enablement</category><category>learning-design</category><category>ai-experiments</category><category>Agent-Ready Knowledge</category></item><item><title>The knowledge nobody has at the moment they need it</title><link>https://bil.arikan.ca/posts/agent-ready-knowledge-nobody-has-at-the-moment-of-need/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/agent-ready-knowledge-nobody-has-at-the-moment-of-need/</guid><description>Tips, documents, onboarding and on-the-job training each cover part of a task. Most enablement still misses the moment of work.</description><pubDate>Thu, 27 Aug 2026 04:00:00 GMT</pubDate><category>learning-design</category><category>performance-support</category><category>sales-enablement</category><category>knowledge-management</category><category>learning-architecture</category><category>Agent-Ready Knowledge</category></item><item><title>Vector, graph, or a curated wiki --- how a purpose-built agent can get its knowledge</title><link>https://bil.arikan.ca/posts/agent-ready-knowledge-vector-graph-or-curated-wiki/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/agent-ready-knowledge-vector-graph-or-curated-wiki/</guid><description>I came to curated knowledge from vector search and direct documents. Here is how I would choose between them for a purpose-built agent.</description><pubDate>Fri, 14 Aug 2026 04:00:00 GMT</pubDate><category>rag</category><category>okf</category><category>llm-wiki</category><category>ai-agents</category><category>knowledge-management</category><category>Agent-Ready Knowledge</category></item><item><title>Markdown files an agent can trust --- an introduction to LLM Wiki and the Open Knowledge Format</title><link>https://bil.arikan.ca/posts/agent-ready-knowledge-llm-wiki-and-open-knowledge-format-introduction/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/agent-ready-knowledge-llm-wiki-and-open-knowledge-format-introduction/</guid><description>A plain-language introduction to the LLM Wiki pattern and the Open Knowledge Format, and how a small agent can use a folder of Markdown files as its knowledge.</description><pubDate>Sat, 01 Aug 2026 04:00:00 GMT</pubDate><category>okf</category><category>llm-wiki</category><category>ai-agents</category><category>knowledge-management</category><category>markdown</category><category>google-adk</category><category>Agent-Ready Knowledge</category></item><item><title>What AI Builders Inside Large Orgs Actually Deal With When They Are Not a Part of Development, Engineering, or Technology Teams</title><link>https://bil.arikan.ca/posts/what-ai-builders-inside-large-orgs-actually-deal-with/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/what-ai-builders-inside-large-orgs-actually-deal-with/</guid><description>Building AI prototypes inside a large enterprise is a political challenge as much as a technical one, and the patterns that sideline good ideas are predictable.</description><pubDate>Sun, 07 Jun 2026 00:00:00 GMT</pubDate><category>ai-builder</category><category>intrapreneurship</category><category>enterprise-ai</category><category>prototyping</category><category>org-dynamics</category><category>innovation</category><category>learning-technologist</category><category>perspectives</category></item><item><title>Code-as-Video for Learning Content</title><link>https://bil.arikan.ca/posts/code-as-video/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/code-as-video/</guid><description>A running experiment in treating learning and product video like code: versioned, reproducible, and re-rendered on demand with an agent in the loop.</description><pubDate>Mon, 01 Jun 2026 05:00:00 GMT</pubDate><category>code-as-video</category><category>editframe</category><category>video-production</category><category>ai-agents</category><category>learning-and-development</category><category>instructional-design</category><category>experiments</category><category>workflow</category></item><item><title>Local AI Lab --- The Full Stack of a Local Claude Cowork and Code Alternative I Have to Build Myself</title><link>https://bil.arikan.ca/posts/local-ai-lab-the-full-stack-reviewed/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-the-full-stack-reviewed/</guid><description>An early, honest review of the finished local stack: how close it gets to Claude Cowork and Claude Code, and where it falls short.</description><pubDate>Sun, 31 May 2026 20:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>Proxmox AI Lab</category></item><item><title>The Proxmox Local AI Lab</title><link>https://bil.arikan.ca/posts/local-ai-lab/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab/</guid><description>The map of the Proxmox AI Lab: local inference on two GPUs, a coding agent, and a gateway reachable from my phone, with no cloud AI in the loop.</description><pubDate>Sun, 31 May 2026 13:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Connecting Telegram to Your Local AI</title><link>https://bil.arikan.ca/posts/local-ai-lab-connecting-telegram-to-your-local-ai/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-connecting-telegram-to-your-local-ai/</guid><description>Pairing a Telegram bot with OpenClaw, and the static IP change that silently killed the connection.</description><pubDate>Sat, 30 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>openclaw</category><category>telegram</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Bridging OpenClaw and Pi Across Containers</title><link>https://bil.arikan.ca/posts/local-ai-lab-bridging-openclaw-and-pi/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-bridging-openclaw-and-pi/</guid><description>Wiring OpenClaw and Pi together across containers with a small HTTP bridge, single-flight queueing, and a shared-secret token.</description><pubDate>Fri, 29 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>pi</category><category>openclaw</category><category>coding-agent</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Setting Up OpenClaw as a Personal AI Gateway</title><link>https://bil.arikan.ca/posts/local-ai-lab-setting-up-openclaw/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-setting-up-openclaw/</guid><description>What an AI gateway actually does, why I picked OpenClaw, and setting it up as the layer that makes local models reachable from my phone.</description><pubDate>Thu, 28 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>openclaw</category><category>telegram</category><category>signal</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Setting Up Pi as a Local Coding Agent</title><link>https://bil.arikan.ca/posts/local-ai-lab-setting-up-pi-coding-agent/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-setting-up-pi-coding-agent/</guid><description>Setting up Pi as a local coding agent against a 27B model on llama-server, from a fresh container to a passing smoke test.</description><pubDate>Thu, 28 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>pi</category><category>coding-agent</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Inference on Two GPUs --- llama.cpp on CUDA and ROCm</title><link>https://bil.arikan.ca/posts/local-ai-lab-inference-on-two-gpus/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-inference-on-two-gpus/</guid><description>Why llama.cpp beat Ollama for this hardware, and building two llama-server containers: one on CUDA, one on ROCm.</description><pubDate>Tue, 26 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Minecraft Server on Proxmox --- LAN Play for Two</title><link>https://bil.arikan.ca/posts/local-ai-lab-minecraft-server/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-minecraft-server/</guid><description>A PaperMC Minecraft server in an LXC container on Proxmox for LAN play between two laptops, gotchas included.</description><pubDate>Mon, 25 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>minecraft</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Designing and Building a 3-Node Proxmox Cluster for AI</title><link>https://bil.arikan.ca/posts/local-ai-lab-proxmox-cluster-design-and-setup/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-proxmox-cluster-design-and-setup/</guid><description>The design decisions and hands-on build of a 3-node Proxmox cluster for local AI inference, homelab services, and a phone-reachable agent stack.</description><pubDate>Sun, 24 May 2026 05:00:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>Proxmox AI Lab</category></item><item><title>Local AI Lab --- Why I Want to Stop Relying on Cloud AI for Everything</title><link>https://bil.arikan.ca/posts/local-ai-lab-why-local-ai-lab/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/local-ai-lab-why-local-ai-lab/</guid><description>After a year of leaning on cloud AI for everything, why a local, always-on AI stack is the next step.</description><pubDate>Thu, 21 May 2026 13:30:00 GMT</pubDate><category>proxmox</category><category>homelab</category><category>ai-infra</category><category>proxmox-ai-lab</category><category>Proxmox AI Lab</category></item><item><title>Setting up a custom ATProto handle with a domain you already own</title><link>https://bil.arikan.ca/posts/custom-atproto-handle/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/custom-atproto-handle/</guid><description>How I moved from @bilarikan.bsky.social to @bil.arikan.ca on Bluesky and Tangled, including the two attempts that failed.</description><pubDate>Mon, 18 May 2026 00:00:00 GMT</pubDate><category>atproto</category><category>bluesky</category><category>tangled</category><category>dns</category><category>hugo</category><category>indieweb</category><category>experiments</category></item><item><title>Claude for Small Business --- two edges of the same toggle</title><link>https://bil.arikan.ca/posts/claude-for-small-business-two-edges/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/claude-for-small-business-two-edges/</guid><description>Claude for Small Business makes being a connector inside Claude essential for SMB software, and puts those same products a step closer to being replaced.</description><pubDate>Thu, 14 May 2026 00:00:00 GMT</pubDate><category>claude</category><category>anthropic</category><category>claude-for-small-business</category><category>claude-cowork</category><category>agentic-workflows</category><category>connectors</category><category>plugins</category><category>skills</category><category>mcp</category><category>smb</category><category>platform-strategy</category><category>learning-technologist</category><category>perspectives</category></item><item><title>Mapping the code-as-video landscape --- deterministic to generative, with a learning-content lens</title><link>https://bil.arikan.ca/posts/code-as-video-landscape-for-learning-content/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/code-as-video-landscape-for-learning-content/</guid><description>A map of the code-as-video field, from Manim and Remotion to Sora, sorted from deterministic to generative by what each tier offers learning content.</description><pubDate>Fri, 08 May 2026 00:00:00 GMT</pubDate><category>code-as-video</category><category>editframe</category><category>hyperframes</category><category>manim</category><category>motion-canvas</category><category>remotion</category><category>revideo</category><category>learning-and-development</category><category>instructional-design</category><category>ai-agents</category><category>landscape</category><category>perspectives</category></item><item><title>Code-as-video with Editframe --- video production as a dev workflow</title><link>https://bil.arikan.ca/posts/code-as-video-with-editframe/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/code-as-video-with-editframe/</guid><description>An agent reads Akaunting&apos;s codebase and produces a 71-second walkthrough, which changes how Learning Program Owners, Designers, and Developers split the work.</description><pubDate>Thu, 07 May 2026 00:00:00 GMT</pubDate><category>editframe</category><category>video-production</category><category>code-as-video</category><category>ai-agents</category><category>learning-and-development</category><category>instructional-design</category><category>akaunting</category><category>experiments</category><category>workflow</category></item><item><title>Hello world with Editframe --- video production as a dev workflow</title><link>https://bil.arikan.ca/posts/video-production-to-dev-workflow-editframe-test-drive/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/video-production-to-dev-workflow-editframe-test-drive/</guid><description>A first experiment in treating video like code: scaffolding an Editframe project, iterating with an AI agent, and rendering a hello-world MP4.</description><pubDate>Tue, 05 May 2026 00:00:00 GMT</pubDate><category>editframe</category><category>video-production</category><category>code-as-video</category><category>ai-agents</category><category>cursor</category><category>claude-code</category><category>experiments</category><category>workflow</category></item><item><title>Nine levels of personal AI maturity, and the cliff at level six</title><link>https://bil.arikan.ca/posts/nine-levels-of-personal-ai-maturity/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/nine-levels-of-personal-ai-maturity/</guid><description>A nine-level model of personal AI maturity, and why most rollouts stall at the cliff between using AI and building with it.</description><pubDate>Sun, 03 May 2026 00:00:00 GMT</pubDate><category>ai</category><category>agentic-ai</category><category>learning-architecture</category><category>l-and-d</category><category>maturity-model</category><category>personal-ai</category><category>orchestration</category><category>learning-technology</category></item><item><title>From blocked to built : decomposing a stuck project with human + AI collaboration</title><link>https://bil.arikan.ca/posts/from-blocked-to-built/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/from-blocked-to-built/</guid><description>The decomposition approach that took a months-stalled training adaptation project to a stakeholder-validated proposal in seven days.</description><pubDate>Sun, 26 Apr 2026 00:00:00 GMT</pubDate><category>ai</category><category>instructional-design</category><category>prompt-engineering</category><category>human-ai-collaboration</category><category>learning-design</category><category>experiments</category></item><item><title>Git Worktrees and Claude Code: Why My Local Files Didn&apos;t Update</title><link>https://bil.arikan.ca/posts/git-worktrees-claude-code/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/git-worktrees-claude-code/</guid><description>Claude Code&apos;s changes were on GitHub but not in my local files, because it works in a git worktree by default.</description><pubDate>Sat, 18 Apr 2026 16:00:00 GMT</pubDate><category>git</category><category>claude-code</category><category>workflow</category><category>knowledge-management</category></item><item><title>Claude Skills and the Copilot Parallel : A Practitioner&apos;s Map</title><link>https://bil.arikan.ca/posts/claude-skills-copilot-mapping/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/claude-skills-copilot-mapping/</guid><description>A workflow that takes an hour in Claude takes three to five in Copilot, and mapping Claude&apos;s skill system onto M365 shows the gap is architectural.</description><pubDate>Sat, 18 Apr 2026 12:00:00 GMT</pubDate><category>ai</category><category>claude</category><category>copilot</category><category>agents</category><category>workflow</category><category>experiments</category></item><item><title>Android Vault Sync Update: Replacing GitSync with Termux</title><link>https://bil.arikan.ca/posts/android-vault-sync-termux/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/android-vault-sync-termux/</guid><description>GitSync broke the git index every time I added a new note on Android, so I replaced it with Termux.</description><pubDate>Sat, 18 Apr 2026 04:00:00 GMT</pubDate><category>obsidian</category><category>git</category><category>github</category><category>android</category><category>termux</category><category>knowledge-management</category><category>workflow</category></item><item><title>In-app Live Assistant : Part 2 --- Building the Screen-Sharing Version</title><link>https://bil.arikan.ca/posts/in-app-live-assistant-part-2-building-the/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/in-app-live-assistant-part-2-building-the/</guid><description>A custom web client that swaps the ADK camera feed for screen capture, so the agent watches the application instead of the user.</description><pubDate>Mon, 06 Apr 2026 05:00:00 GMT</pubDate><category>google-adk</category><category>vertex-ai</category><category>gemini</category><category>javascript</category><category>screen-sharing</category><category>websocket</category><category>multimodal</category><category>ai-agents</category><category>ai-experiments</category></item><item><title>Translation vs. Localisation Is the Wrong Frame</title><link>https://bil.arikan.ca/posts/translations-localizations-decomposition/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/translations-localizations-decomposition/</guid><description>Arguing over translation versus localisation misses the point; thinking in layers of context maps cleanly onto a multi-agent architecture.</description><pubDate>Thu, 02 Apr 2026 00:00:00 GMT</pubDate><category>localisation</category><category>translation</category><category>ai-agents</category><category>copilot-studio</category><category>learning-design</category><category>instructional-design</category><category>multi-agent</category><category>ai-experiments</category><category>learning-architecture</category></item><item><title>Building a Localisation Agent in Microsoft Copilot Studio</title><link>https://bil.arikan.ca/posts/building-a-localisation-agent-in-copilot-studio/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/building-a-localisation-agent-in-copilot-studio/</guid><description>Instead of buying an expensive translation tool, I built an orchestrated localisation agent in Copilot Studio that never leaves the Microsoft 365 tenant.</description><pubDate>Wed, 18 Mar 2026 05:00:00 GMT</pubDate><category>copilot-studio</category><category>ai-agents</category><category>localisation</category><category>microsoft-365</category><category>anthropic</category><category>ai-experiments</category></item><item><title>In-app Live Assistant : Part 1 --- Walking Through Google&apos;s ADK Bidirectional Streaming Demo</title><link>https://bil.arikan.ca/posts/in-app-live-assistant-part-1-google-adk-bidirectional-streaming-demo/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/in-app-live-assistant-part-1-google-adk-bidirectional-streaming-demo/</guid><description>Getting Google&apos;s ADK bidirectional streaming demo running locally with mic, camera, and voice, as the foundation for a live in-app assistant.</description><pubDate>Mon, 09 Mar 2026 05:00:00 GMT</pubDate><category>google-adk</category><category>bidirectional-streaming</category><category>vertex-ai</category><category>gemini</category><category>python</category><category>multimodal</category><category>ai-agents</category><category>ai-experiments</category></item><item><title>Google Is Quietly Building the Agentic Web</title><link>https://bil.arikan.ca/posts/google-quietly-building-agentic-web/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/google-quietly-building-agentic-web/</guid><description>While OpenAI and Anthropic fought over a Pentagon deal, Google shipped WebMCP, UCP, and A2A, the protocol stack for how agents will use the web.</description><pubDate>Fri, 06 Mar 2026 05:00:00 GMT</pubDate><category>ai-agents</category><category>agentic-web</category><category>google</category><category>protocols</category><category>webmcp</category><category>commerce</category><category>a2a</category><category>ai-experiments</category></item><item><title>Private Obsidian Vault Sync with GitHub</title><link>https://bil.arikan.ca/posts/obsidian-private-github-sync/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/obsidian-private-github-sync/</guid><description>Syncing an Obsidian vault to a private GitHub repo with the Obsidian Git plugin on macOS, with a short Windows addendum.</description><pubDate>Sun, 22 Feb 2026 05:00:00 GMT</pubDate><category>obsidian</category><category>git</category><category>github</category><category>macos</category><category>knowledge-management</category><category>workflow</category></item><item><title>Building Learning Content for Humans and Agents</title><link>https://bil.arikan.ca/posts/future-ready-learning-content/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/future-ready-learning-content/</guid><description>A working model for designing learning content that supports both human understanding and agent retrieval.</description><pubDate>Fri, 13 Feb 2026 05:00:00 GMT</pubDate><category>learning-design</category><category>ai-agents</category><category>knowledge-management</category><category>markdown</category><category>learning-architecture</category></item><item><title>In-App Live Assistant</title><link>https://bil.arikan.ca/posts/live-in-app-assistant/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/live-in-app-assistant/</guid><description>An open-source kit for a real-time in-app assistant that listens to voice, sees the screen, and guides users step by step.</description><pubDate>Fri, 13 Feb 2026 05:00:00 GMT</pubDate><category>open-source</category><category>ai-agents</category><category>multimodal</category><category>streaming</category><category>voice</category><category>screen-sharing</category><category>xapi</category><category>learning-measurement</category></item><item><title>Experiment: A Live In-App Assistant (Voice + Screen)</title><link>https://bil.arikan.ca/posts/live-streaming-in-app-assistant/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/live-streaming-in-app-assistant/</guid><description>Working notes from recreating a voice-and-screen assistant: one path failed on authentication, the next produced a working streaming prototype.</description><pubDate>Fri, 13 Feb 2026 05:00:00 GMT</pubDate><category>ai-agents</category><category>multimodal</category><category>streaming</category><category>voice</category><category>prototyping</category><category>customer-support</category><category>ai-experiments</category></item><item><title>bil.arikan.ca Migration</title><link>https://bil.arikan.ca/posts/site-migration/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/site-migration/</guid><description>How bil.arikan.ca moved from Astro to Hugo with the Congo theme, including the GitHub Pages deployment changes.</description><pubDate>Wed, 11 Feb 2026 05:00:00 GMT</pubDate><category>hugo</category><category>astro</category><category>migration</category><category>github-pages</category><category>congo</category></item><item><title>Welcome: This Site as a Learning Lab</title><link>https://bil.arikan.ca/posts/welcome-to-hugo/</link><guid isPermaLink="true">https://bil.arikan.ca/posts/welcome-to-hugo/</guid><description>Why this site exists: a place to practice learning design, knowledge systems, and pair-programming with AI agents in public.</description><pubDate>Wed, 11 Feb 2026 05:00:00 GMT</pubDate><category>hugo</category><category>congo</category><category>learning</category><category>knowledge-management</category><category>ai-agents</category><category>pair-programming</category><category>website</category></item></channel></rss>