Architecting secure, governed AI at enterprise scale. Essays on AI strategy, governance, and engineering from Andreas Wasita.https://wasita.net/en-au© 2026 Andreas Wasitahttps://wasita.net/blog/copilot-agents-dojo-improvements/https://wasita.net/blog/copilot-agents-dojo-improvements/I shipped a governance framework for AI agents, then failed its own adoption test — no uninstall, no way to list its skills, no way to know if it had drifted. Here's the sprint that fixed it, and the four patterns you can steal whether or not you ever touch my repo.Mon, 08 Jun 2026 00:00:00 GMTGitHub CopilotAI AgentsAgent GovernanceDeveloper ExperienceAndreas Wasitahttps://wasita.net/blog/inside-my-ai-operating-system-part-2/https://wasita.net/blog/inside-my-ai-operating-system-part-2/My 3D AI office lied to me, and the afternoon I lost to it taught me more about governing agents than any amount of infrastructure did. Part II of the AI OS deep dive: telling a dashboard from a trigger, a leash on autonomy you can actually verify, and giving memory tiers.Sat, 06 Jun 2026 00:00:00 GMTAI AgentsHermesMemoryGovernanceObsidianAndreas Wasitahttps://wasita.net/blog/inside-my-ai-operating-system/https://wasita.net/blog/inside-my-ai-operating-system/A technical deep dive into the always-on agent stack that runs my work: a Hermes kernel on my Mac, a Paperclip workforce on a VPS, one Obsidian vault as shared memory, MCP as the syscall layer — and a Tailscale mesh holding two machines together with no open ports.Thu, 04 Jun 2026 00:00:00 GMTAI AgentsMCPObsidianHermesTailscaleAndreas Wasitahttps://wasita.net/blog/my-new-operating-system-hermes-paperclip-obsidian-mcp/https://wasita.net/blog/my-new-operating-system-hermes-paperclip-obsidian-mcp/I stopped thinking of my AI tools as separate apps and started running them like an operating system. Hermes is the always-on kernel, Paperclip is the agent workforce, a Jarvis wake-word loop is the microphone, one Obsidian vault is shared memory for every runtime, and MCP is the syscall layer.Sun, 31 May 2026 00:00:00 GMTAI AgentsMCPObsidianHermesProductivityAndreas Wasitahttps://wasita.net/blog/red-thread-togaf-traceability/https://wasita.net/blog/red-thread-togaf-traceability/Agentic delivery generates plausible artifacts at every architecture layer with no enforced lineage. Here's how to keep the TOGAF red thread unbroken when agents are doing the work.Wed, 27 May 2026 00:00:00 GMTEnterprise ArchitectureTOGAFAgentic AIAgent GovernanceGitHub CopilotAndreas Wasitahttps://wasita.net/blog/evolution-of-skills-md/https://wasita.net/blog/evolution-of-skills-md/Persona libraries, self-improving runtimes, and behavioural governance are three layers of the same stack. The frontier is making them work together.Tue, 26 May 2026 00:00:00 GMTAgentic AIGitHub CopilotAgent GovernanceDeveloper ExperienceAndreas Wasitahttps://wasita.net/blog/spec-kit-togaf-architect-playbook/https://wasita.net/blog/spec-kit-togaf-architect-playbook/Spec-Kit gives AI agents a disciplined workflow. TOGAF gives the enterprise a disciplined architecture. Map them together and you get governed, AI-native delivery.Fri, 22 May 2026 00:00:00 GMTSpec-Driven DevelopmentTOGAFEnterprise ArchitectureGitHub CopilotAndreas Wasitahttps://wasita.net/blog/untrained-agents/https://wasita.net/blog/untrained-agents/We keep waiting for smarter models. But the agents we already have fail for the same reasons junior engineers do — no plan, no proof, no memory. Capability isn't the constraint. Discipline is.Fri, 15 May 2026 00:00:00 GMTAgentic AIEngineering practiceCopilot Agents DojoAndreas Wasitahttps://wasita.net/blog/mandatory-pipeline/https://wasita.net/blog/mandatory-pipeline/Conventional wisdom says you give a capable agent room to work. I did the opposite: a fixed, non-negotiable workflow from brainstorm to finish. Constraint didn't slow the agents down. It's what made them trustworthy.Sun, 10 May 2026 00:00:00 GMTEngineering practiceAgentic AICopilot Agents DojoAndreas Wasitahttps://wasita.net/blog/learning-from-losing/https://wasita.net/blog/learning-from-losing/Most agent setups make the same mistake twice — or twenty times. The most valuable thing I built into the dojo wasn't a skill. It was a loop that turns every correction into a rule the agent can't forget.Tue, 05 May 2026 00:00:00 GMTSelf-improving systemsAgentic AICopilot Agents DojoAndreas Wasitahttps://wasita.net/blog/building-lawyer-companion/https://wasita.net/blog/building-lawyer-companion/How I transformed a Next.js landing page into a full serverless SaaS with Document Intelligence and Chat Your Data — in a single Copilot session.Sun, 26 Apr 2026 00:00:00 GMTGitHub CopilotServerlessAWSBuild LogAndreas Wasitahttps://wasita.net/blog/claude-vs-gpt-enterprise/https://wasita.net/blog/claude-vs-gpt-enterprise/A practitioner's honest comparison of Claude and GPT models in enterprise settings — strengths, trade-offs, and when to use which.Sat, 18 Apr 2026 00:00:00 GMTAnthropic ClaudeAzure OpenAIEnterprise AIAndreas Wasitahttps://wasita.net/blog/azure-ai-foundry-production-patterns/https://wasita.net/blog/azure-ai-foundry-production-patterns/Practical patterns for deploying AI models in production using Azure AI Foundry — from model selection to cost optimization.Sun, 12 Apr 2026 00:00:00 GMTAzure AIAI FoundryProductionAndreas Wasitahttps://wasita.net/blog/ai-delivery-methodology/https://wasita.net/blog/ai-delivery-methodology/Agile, Scrum, and waterfall weren't designed for AI-assisted development. We need an AI-native delivery methodology.Sun, 05 Apr 2026 00:00:00 GMTAI StrategyEnterpriseMethodologyAndreas Wasitahttps://wasita.net/blog/copilot-agents-dojo-framework/https://wasita.net/blog/copilot-agents-dojo-framework/Most organisations let AI agents loose with prompts and hope for the best. That's not an operating model — that's a risk. The Dojo changes that.Sat, 28 Mar 2026 00:00:00 GMTGitHub CopilotAgent GovernanceOpen SourceAndreas Wasitahttps://wasita.net/blog/governing-ai-agents-at-scale/https://wasita.net/blog/governing-ai-agents-at-scale/If your AI coding strategy still relies on prompts, you're leaving leverage on the table. Here's how top teams govern AI agent behavior at the repo level.Wed, 25 Mar 2026 00:00:00 GMTGitHub CopilotAgentic AIEngineering LeadershipAndreas Wasita