<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>McIntosh Consulting — Insights</title><description>Notes on biotech R&amp;D, AI engineering, and the seams where they meet — from McIntosh Consulting.</description><link>https://vernonmcintosh.com/</link><language>en-us</language><item><title>Five Models, One Spec: Notes from Routing the Same Work to Different LLMs</title><link>https://vernonmcintosh.com/blog/multi-model-llm-dispatch/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/multi-model-llm-dispatch/</guid><description>When you dispatch the same tightly-specified task to GPT-5, GLM-5.1, GLM-4.6, and DeepSeek-V3 in parallel, the failure modes are model-specific in ways that matter. A working set of rules for choosing the right model per task.</description><pubDate>Mon, 11 May 2026 00:00:00 GMT</pubDate><category>AI Engineering</category></item><item><title>The Hidden Tax on AI Agent Swarms</title><link>https://vernonmcintosh.com/blog/hidden-cost-agent-swarms/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/hidden-cost-agent-swarms/</guid><description>Decomposing work across multiple AI agents often costs more than a single well-designed agent — here is why that happens and what the numbers actually look like.</description><pubDate>Sat, 09 May 2026 00:00:00 GMT</pubDate><category>AI Architecture</category></item><item><title>When Your Fermentation Data Talks Back: AI-Augmented Bioprocess Development</title><link>https://vernonmcintosh.com/blog/ai-augmented-bioprocess/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/ai-augmented-bioprocess/</guid><description>AI and ML are shifting bioprocess development from trial-and-error to data-driven optimization by giving scientists pattern recognition across thousands of fermentation runs that no human can replicate manually.</description><pubDate>Fri, 10 Apr 2026 00:00:00 GMT</pubDate><category>Biotech</category></item><item><title>The Synthetic Biology Stack: Why Biotech Needs Software Engineering Principles</title><link>https://vernonmcintosh.com/blog/synthetic-biology-stack/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/synthetic-biology-stack/</guid><description>Biotech companies that adopt software engineering practices (version control for strains, CI/CD for genetic constructs, automated testing for phenotypes) will outcompete those that don&apos;t.</description><pubDate>Wed, 18 Mar 2026 00:00:00 GMT</pubDate><category>Biotech</category></item><item><title>Validating Synthetic Genomic Data: The Missing Quality Layer</title><link>https://vernonmcintosh.com/blog/validating-synthetic-genomics/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/validating-synthetic-genomics/</guid><description>As synthetic genomic data becomes critical for ML training and privacy-preserving research, we need validation frameworks that measure fidelity, utility, and privacy, not just statistical similarity.</description><pubDate>Tue, 24 Feb 2026 00:00:00 GMT</pubDate><category>Data Science</category></item><item><title>The Git Repo as the AI Workflow System Boundary</title><link>https://vernonmcintosh.com/blog/git-repo-system-boundary/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/git-repo-system-boundary/</guid><description>A git repository is the natural and complete system boundary for AI-driven document workflows. Everything the agent needs already lives in or can be referenced from a repo.</description><pubDate>Tue, 03 Feb 2026 00:00:00 GMT</pubDate><category>AI Architecture</category></item><item><title>What Is the Smallest Unit of Work in Agent Decomposition?</title><link>https://vernonmcintosh.com/blog/smallest-unit-of-work/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/smallest-unit-of-work/</guid><description>When decomposing tasks for AI agents, there&apos;s a granularity floor: a point below which further decomposition creates more overhead than value.</description><pubDate>Wed, 14 Jan 2026 00:00:00 GMT</pubDate><category>AI Architecture</category></item><item><title>Knowing That You Know: External Memory Architecture for AI</title><link>https://vernonmcintosh.com/blog/knowing-that-you-know/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/knowing-that-you-know/</guid><description>Most AI memory architectures have two states: knowing and not knowing. A third state, knowing that you know, lets systems scale beyond their context window without loading everything at once.</description><pubDate>Fri, 19 Dec 2025 00:00:00 GMT</pubDate><category>AI Architecture</category></item><item><title>Innovation Driven by Ability, Not Self-Awareness</title><link>https://vernonmcintosh.com/blog/innovation-driven-by-ability/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/innovation-driven-by-ability/</guid><description>Capability expansion precedes need articulation. The most consequential technologies in history weren&apos;t solutions to articulated problems.</description><pubDate>Tue, 02 Dec 2025 00:00:00 GMT</pubDate><category>R&amp;D Strategy</category></item><item><title>Starting With What&apos;s Possible: Why R&amp;D Should Lead Innovation</title><link>https://vernonmcintosh.com/blog/rnd-first-bullseye/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/rnd-first-bullseye/</guid><description>The R&amp;D-First Bullseye Model reverses traditional market-led approaches by starting with technical possibility and working outward toward market validation.</description><pubDate>Tue, 11 Nov 2025 00:00:00 GMT</pubDate><category>Innovation</category></item><item><title>The Gold Mine in Your Lost Proposals</title><link>https://vernonmcintosh.com/blog/gold-mine-lost-proposals/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/gold-mine-lost-proposals/</guid><description>Your organization&apos;s lost proposals and unpursued opportunities aren&apos;t failures. They&apos;re an asset library waiting to be indexed.</description><pubDate>Tue, 21 Oct 2025 00:00:00 GMT</pubDate><category>Business Development</category></item><item><title>The Context Window is a Viewport, Not a Bucket</title><link>https://vernonmcintosh.com/blog/context-window-viewport/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/context-window-viewport/</guid><description>Knowledge map traversal (navigating structured graphs on-demand instead of loading entire databases) is the superior pattern regardless of context window size.</description><pubDate>Mon, 29 Sep 2025 00:00:00 GMT</pubDate><category>AI Architecture</category></item><item><title>The Syntax Tax</title><link>https://vernonmcintosh.com/blog/democratization-salesforce-dev/</link><guid isPermaLink="true">https://vernonmcintosh.com/blog/democratization-salesforce-dev/</guid><description>Enterprise and scientific software has a long tail of capability most teams never reach because it is locked behind syntax. AI coding assistants change where in the workflow that translation happens.</description><pubDate>Mon, 08 Sep 2025 00:00:00 GMT</pubDate><category>AI + Tools</category></item></channel></rss>