{"@attributes":{"version":"2.0"},"channel":{"title":"mick delaney","description":"I'm a CTO based in London, UK. I love building products and teams that solve real problems. I write about software engineering, product management, and leadership.","link":"https:\/\/mickdelaney.com\/","item":[{"title":"Why Agents Make Good Actors","link":"https:\/\/mickdelaney.com\/posts\/why-agents-make-good-actors\/","guid":"https:\/\/mickdelaney.com\/posts\/why-agents-make-good-actors\/","description":"AI agents are stateful, long-running, and mostly idle. That's not a quirk \u2014 it's a specification. The actor model was built for exactly this shape of work.","pubDate":"Sun, 28 Jun 2026 10:00:00 GMT"},{"title":"An introduction to reinforcement learning","link":"https:\/\/mickdelaney.com\/posts\/rl-1-introduction\/","guid":"https:\/\/mickdelaney.com\/posts\/rl-1-introduction\/","description":"The orientation for working engineers before the rest of the series goes into algorithm internals. What RL is, the vocabulary, the Bellman idea, when to reach for it, and a map of the five posts.","pubDate":"Wed, 03 Jun 2026 10:00:00 GMT"},{"title":"Designing environments for RL","link":"https:\/\/mickdelaney.com\/posts\/rl-2-designing-environments\/","guid":"https:\/\/mickdelaney.com\/posts\/rl-2-designing-environments\/","description":"Most RL writing is about algorithms. In practice, environment design \u2014 the state, action, and reward decisions that sit between your system and the algorithm \u2014 is where projects succeed or fail. This is the part that almost never gets covered.","pubDate":"Wed, 03 Jun 2026 10:00:00 GMT"},{"title":"Inside the Bellman loop: how value actually propagates","link":"https:\/\/mickdelaney.com\/posts\/rl-3-inside-the-bellman-loop\/","guid":"https:\/\/mickdelaney.com\/posts\/rl-3-inside-the-bellman-loop\/","description":"Bellman equations are usually introduced recursively, and the recursion misleads programmers. The runtime is closer to iterative graph relaxation \u2014 value sweeping backward through a state graph until it stops moving.","pubDate":"Wed, 27 May 2026 18:00:00 GMT"},{"title":"Hardening AI Workflows with the Agent Governance Framework","link":"https:\/\/mickdelaney.com\/posts\/agent-governance-maf-recruitment\/","guid":"https:\/\/mickdelaney.com\/posts\/agent-governance-maf-recruitment\/","description":"How to use the Microsoft Agent Framework (MAF) and Agent Governance Toolkit to build secure, governed, and Human-in-the-Loop AI workflows.","pubDate":"Sun, 03 May 2026 10:00:00 GMT"},{"title":"Securing MCP Pipelines with Agent Governance","link":"https:\/\/mickdelaney.com\/posts\/agent-governance-mcp-pipelines\/","guid":"https:\/\/mickdelaney.com\/posts\/agent-governance-mcp-pipelines\/","description":"How to secure Model Context Protocol (MCP) tool usage in sequential agent workflows using Supply Chain Validation and Least-Privilege Sandboxing.","pubDate":"Sun, 03 May 2026 10:00:00 GMT"},{"title":"Inside Microsoft Agent Framework Harness Providers","link":"https:\/\/mickdelaney.com\/posts\/maf-harness-providers\/","guid":"https:\/\/mickdelaney.com\/posts\/maf-harness-providers\/","description":"An LLM in a loop is not enough for production work. This post breaks down the six problems every agent harness must solve \u2014 working memory, shared file access, plan\/execute modes, task tracking, parallel sub-agents, and tool approval \u2014 and how the Microsoft Agents Framework implements them.","pubDate":"Sat, 02 May 2026 10:00:00 GMT"},{"title":"Securing Agent-to-Agent (A2A) Handoffs with the Agent Mesh","link":"https:\/\/mickdelaney.com\/posts\/agent-governance-a2a-handoffs\/","guid":"https:\/\/mickdelaney.com\/posts\/agent-governance-a2a-handoffs\/","description":"How to secure distributed multi-agent architectures using Zero-Trust Identity, IATP encryption, and Dynamic Trust Scoring in the Agent Governance Toolkit.","pubDate":"Fri, 01 May 2026 14:00:00 GMT"},{"title":"Inside the ReAct loop: how agents actually iterate","link":"https:\/\/mickdelaney.com\/posts\/inside-the-react-loop\/","guid":"https:\/\/mickdelaney.com\/posts\/inside-the-react-loop\/","description":"What actually happens inside one iteration of the ReAct loop \u2014 why the model is stateless, how the transcript grows, when the loop exits, and why the harness around the model is where the engineering value lives.","pubDate":"Mon, 20 Apr 2026 10:00:00 GMT"},{"title":"The Protocol Layer: Watching the Theory Play Out in Real Time","link":"https:\/\/mickdelaney.com\/posts\/the-protocol-layer-watching-theory-play-out\/","guid":"https:\/\/mickdelaney.com\/posts\/the-protocol-layer-watching-theory-play-out\/","description":"Google's developer guide to AI agent protocols is framed as a tutorial. Read it through a Coasean lens and it's something more interesting \u2014 live evidence that the transaction costs making firms necessary are being eliminated, layer by layer, in the open.","pubDate":"Fri, 03 Apr 2026 10:00:00 GMT"},{"title":"The Firm in 2026","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-16-the-firm-in-2026\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-16-the-firm-in-2026\/","description":"Returning to Coase's original question \u2014 why do firms exist? \u2014 with fifteen posts of new machinery. The answer has not changed. The composition of the costs has.","pubDate":"Tue, 17 Mar 2026 10:00:00 GMT"},{"title":"The Agentic Ecosystem","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-15-the-agentic-ecosystem\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-15-the-agentic-ecosystem\/","description":"The shift from hierarchy to protocol-governed agent coordination is not theoretical. Over fifty companies are engineering it now. Here is what the early evidence tells us.","pubDate":"Sun, 15 Mar 2026 10:00:00 GMT"},{"title":"Value Stream Mapping","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-14-value-stream-mapping\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-14-value-stream-mapping\/","description":"Protocols define how participants interact. Value streams define why. By mapping the flow of outcomes rather than the structure of teams, we align architecture, organisation, and network around the same thing: how value is actually produced.","pubDate":"Fri, 13 Mar 2026 07:00:00 GMT"},{"title":"The Physics of Flow","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-13-the-physics-of-flow\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-13-the-physics-of-flow\/","description":"Queuing theory reveals the hidden physics governing how work flows through organisations. Every boundary is a queue. High utilisation destroys responsiveness. And AI agents change the service rate \u2014 but not the dynamics.","pubDate":"Thu, 12 Mar 2026 10:00:00 GMT"},{"title":"The Protocol Layer","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-12-the-protocol-layer\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-12-the-protocol-layer\/","description":"How networked firms coordinate work through protocols instead of hierarchy.","pubDate":"Wed, 11 Mar 2026 10:00:00 GMT"},{"title":"The Geometry of Networks","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-11-the-geometry-of-networks\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-11-the-geometry-of-networks\/","description":"Not all networks scale the same way. The geometry of a network \u2014 broadcast, connection, or group-forming \u2014 determines how its value grows, and which institutions it produces.","pubDate":"Mon, 09 Mar 2026 12:00:00 GMT"},{"title":"Metcalfe's Law, Conway's Law, and the Networked Firm","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-10-metcalfes-law-conways-law-networked-firm\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-10-metcalfes-law-conways-law-networked-firm\/","description":"Metcalfe's Law governs the value of the external network, while Conway's Law governs the structure of the organization that builds and operates it.","pubDate":"Sun, 08 Mar 2026 10:00:00 GMT"},{"title":"The Hybrid Topology","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-9-the-hybrid-topology\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-9-the-hybrid-topology\/","description":"We are witnessing the collapse of the corporate Tree into the Hybrid Mesh. As intelligence becomes infrastructure, the distance between strategy and execution shrinks to a single, high-density hop.","pubDate":"Sat, 07 Mar 2026 10:00:00 GMT"},{"title":"Knowledge Diffusion and Network Density","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-8-knowledge-diffusion-and-network-density\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-8-knowledge-diffusion-and-network-density\/","description":"Firms do not only coordinate production. They concentrate knowledge. Innovation has historically depended on short-distance diffusion through dense networks.","pubDate":"Sun, 22 Feb 2026 10:00:00 GMT"},{"title":"Agentic Labour: When Work Becomes Executable","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-7-agentic-labour-when-work-becomes-executable\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-7-agentic-labour-when-work-becomes-executable\/","description":"Software first reduced coordination costs. Now autonomous agents begin to execute parts of labor itself. This changes the structure of work without eliminating it.","pubDate":"Fri, 20 Feb 2026 18:30:00 GMT"},{"title":"The Platform as Proto-Firm","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-6-the-platform-as-proto-firm\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-6-the-platform-as-proto-firm\/","description":"Platforms are not the disappearance of the corporation. They are a reconfiguration of how coordination occurs within a corporate shell.","pubDate":"Thu, 19 Feb 2026 20:15:00 GMT"},{"title":"The Boundary of the Firm in a Digital Age","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-5-the-boundary-of-the-firm-in-a-digital-age\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-5-the-boundary-of-the-firm-in-a-digital-age\/","description":"If firms exist because coordination is costly, what happens when coordination becomes cheaper again? The boundary of the firm is not fixed.","pubDate":"Wed, 18 Feb 2026 18:30:00 GMT"},{"title":"The Innovator's Dilemma","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-4-the-innovators-dilemma\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-4-the-innovators-dilemma\/","description":"When coordination costs shift, firm boundaries should adjust. But mature corporations are optimised for a cost structure that no longer holds \u2014 and their own incentives prevent adaptation.","pubDate":"Tue, 17 Feb 2026 20:15:00 GMT"},{"title":"When Corporations Became Too Big","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-3-when-corporations-became-too-big\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-3-when-corporations-became-too-big\/","description":"The managerial corporation solved coordination at scale. But hierarchy has structural limits \u2014 and the same mechanisms that enabled growth eventually constrained it.","pubDate":"Mon, 16 Feb 2026 19:15:00 GMT"},{"title":"The Rise of the Managerial Corporation","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-2-rise-of-the-managerial-corporation\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-2-rise-of-the-managerial-corporation\/","description":"Corporations did not just reduce transaction costs. In the 20th century, they became the dominant institution for organising production, concentrating capital, and coordinating labour at scale.","pubDate":"Sun, 15 Feb 2026 18:30:00 GMT"},{"title":"Why Corporations Exist: A Coasean Foundation","link":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-1-why-corporations-exist-a-coasean-foundation\/","guid":"https:\/\/mickdelaney.com\/posts\/firm-under-ai-1-why-corporations-exist-a-coasean-foundation\/","description":"Corporations did not emerge by accident. They are coordination machines. In 2026, as transaction costs fall, we revisit why they exist \u2014 and what might be changing.","pubDate":"Sat, 14 Feb 2026 10:00:00 GMT"},{"title":"Metrics Are Projection","link":"https:\/\/mickdelaney.com\/posts\/metrics-are-projection\/","guid":"https:\/\/mickdelaney.com\/posts\/metrics-are-projection\/","description":"Metrics don't measure reality \u2014 they project it. As systems become more autonomous, projection design matters more than model intelligence.","pubDate":"Mon, 09 Feb 2026 10:00:00 GMT"},{"title":"Agent Experimentation and Innovation at the Edge","link":"https:\/\/mickdelaney.com\/posts\/agent-experimentation-at-the-edge\/","guid":"https:\/\/mickdelaney.com\/posts\/agent-experimentation-at-the-edge\/","description":"Execution, collaboration, and ecosystems: three edges where agent innovation is actually happening \u2014 and what it means for building durable, human-centered systems.","pubDate":"Wed, 04 Feb 2026 10:00:00 GMT"},{"title":"From Process Managers to Stable Agent Workflows","link":"https:\/\/mickdelaney.com\/posts\/agent-stability\/","guid":"https:\/\/mickdelaney.com\/posts\/agent-stability\/","description":"Agent workflows feel new, but the stability problems they face are not. Here's how decades of integration architecture apply to building production agent systems.","pubDate":"Sat, 24 Jan 2026 10:00:00 GMT"},{"title":"Keeping State Consistent: Database Transactions in LangGraph Workflows","link":"https:\/\/mickdelaney.com\/posts\/langgraph-database-transactions\/","guid":"https:\/\/mickdelaney.com\/posts\/langgraph-database-transactions\/","description":"Three patterns for keeping LangGraph workflow state and database state synchronized - idempotent tasks, read-before-write synchronization, and the transactional outbox pattern.","pubDate":"Thu, 22 Jan 2026 18:00:00 GMT"},{"title":"An Intuitive Journey Through Statistics & Hypothesis Testing","link":"https:\/\/mickdelaney.com\/posts\/data-to-decisions\/","guid":"https:\/\/mickdelaney.com\/posts\/data-to-decisions\/","description":"An Intuitive Journey Through Statistics & Hypothesis Testing in a coffee shop","pubDate":"Mon, 17 Nov 2025 10:00:00 GMT"},{"title":"Maths-as-Code for Machine Learning","link":"https:\/\/mickdelaney.com\/posts\/maths-as-code-for-ml\/","guid":"https:\/\/mickdelaney.com\/posts\/maths-as-code-for-ml\/","description":"Translate the maths you see in ML papers into code you can write. A compact cheat sheet mapping notation \u2192 intuition \u2192 Python\/NumPy (and JS).","pubDate":"Mon, 03 Nov 2025 00:00:00 GMT"},{"title":"Agents, Routing, Patterns, and Actors","link":"https:\/\/mickdelaney.com\/posts\/agents-routing-patterns-and-actors\/","guid":"https:\/\/mickdelaney.com\/posts\/agents-routing-patterns-and-actors\/","description":"Building Agent Networks that Don't Collapse Under Their Own Cleverness","pubDate":"Tue, 17 Jun 2025 10:00:00 GMT"},{"title":"Agents Are Still Just Software","link":"https:\/\/mickdelaney.com\/posts\/agents-are-still-just-software\/","guid":"https:\/\/mickdelaney.com\/posts\/agents-are-still-just-software\/","description":"Agents Are Just Software Why the Next Wave of AI Is Really a Distributed-Systems Story.","pubDate":"Sun, 08 Jun 2025 10:00:00 GMT"}]}}