<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>AI Colleagues — Blog</title><description>Engineering notes and implementation practices from AI Colleagues.</description><link>https://ai-colleagues.com/</link><item><title>Dev Log: Building the Orcheo Plugin System</title><link>https://ai-colleagues.com/blog/orcheo-plugin-system-dev-log/</link><guid isPermaLink="true">https://ai-colleagues.com/blog/orcheo-plugin-system-dev-log/</guid><description>An inside look at the architecture decisions, implementation choices, and lessons learned while building Orcheo&apos;s managed plugin system — covering the loader, lifecycle CLI, stack-aware runtime targeting, the PluginAPI contract, and the WeCom and Lark listener plugins that validated the design.</description><pubDate>Fri, 20 Mar 2026 00:00:00 GMT</pubDate><category>orcheo</category><category>plugins</category><category>architecture</category><category>developer-experience</category><category>cli</category><category>listeners</category><category>ai-agents</category></item><item><title>Orcheo Quick Start: From Your First Live Workflow to Private AI Bots</title><link>https://ai-colleagues.com/blog/orcheo-local-server-quick-start-from-install-to-your-first-live-workflow/</link><guid isPermaLink="true">https://ai-colleagues.com/blog/orcheo-local-server-quick-start-from-install-to-your-first-live-workflow/</guid><description>A practical introduction to Orcheo for first-time users: install the platform, start from Canvas templates, vibe-code a new workflow, deploy it to an Orcheo local server on your laptop, and then extend the same ideas into private AI bot workflows across Telegram, Discord, and QQ.</description><pubDate>Thu, 12 Mar 2026 00:00:00 GMT</pubDate><category>orcheo</category><category>quick-start</category><category>canvas</category><category>workflow-automation</category><category>vibe-coding</category><category>credential-vault</category><category>telegram</category><category>discord</category><category>qq</category><category>ai-agents</category></item><item><title>Dev Log: Why Orcheo Canvas Is Embracing Vibe-Coded Python Workflows</title><link>https://ai-colleagues.com/blog/dev-log-why-orcheo-canvas-is-embracing-vibe-coded-python-workflows/</link><guid isPermaLink="true">https://ai-colleagues.com/blog/dev-log-why-orcheo-canvas-is-embracing-vibe-coded-python-workflows/</guid><description>We are shifting Canvas workflow authoring from drag-and-drop editing to vibe-coding in Python: why we are making this change now, what we learned from building the visual editor, and how visualisation still matters.</description><pubDate>Fri, 20 Feb 2026 00:00:00 GMT</pubDate><category>orcheo</category><category>canvas</category><category>workflow-authoring</category><category>vibe-coding</category><category>python</category><category>developer-experience</category><category>langgraph</category><category>ai-agents</category></item><item><title>Applying Orcheo to Conversational Search: Key Insights from Our New Paper</title><link>https://ai-colleagues.com/blog/applying-orcheo-to-conversational-search-key-insights-from-our-new-paper/</link><guid isPermaLink="true">https://ai-colleagues.com/blog/applying-orcheo-to-conversational-search-key-insights-from-our-new-paper/</guid><description>Key takeaways from our new paper on building conversational search systems with Orcheo: modular workflows, reproducible experimentation, and a smoother path from prototype to production.</description><pubDate>Mon, 16 Feb 2026 00:00:00 GMT</pubDate><category>orcheo</category><category>conversational-search</category><category>architecture</category><category>reproducibility</category><category>research-paper</category></item></channel></rss>