<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Nikolaj Fløjgaard</title><description>Senior Integration Architect writing about AI agents, systems, autonomy, and the infrastructure behind what actually works.</description><link>https://nikolajflojgaard.me/</link><item><title>The point of an MCP platform is not more code</title><link>https://nikolajflojgaard.me/posts/2026/2026-08-13-building-mcp-platform-limited-budget/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-08-13-building-mcp-platform-limited-budget/</guid><description>AI can generate more code. That is not the bottleneck. The real work is removing integration, ownership, approval, and governance bottlenecks with a domain-driven MCP platform, APIM, and MLflow from mlflow.org.</description><pubDate>Thu, 13 Aug 2026 17:30:00 GMT</pubDate></item><item><title>I made my agent&apos;s memory faster by making it smaller</title><link>https://nikolajflojgaard.me/posts/2026/2026-08-07-memory-cluster-index-agent-memory/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-08-07-memory-cluster-index-agent-memory/</guid><description>The useful version of agent memory is not a bigger bucket. It is a layered system: compact truth, raw notes, topic summaries, and a local index that works even when embeddings do not.</description><pubDate>Fri, 07 Aug 2026 10:58:00 GMT</pubDate></item><item><title>Daniel Kokotajlo is probably right, just too early</title><link>https://nikolajflojgaard.me/posts/2026/2026-07-29-daniel-kokotajlo-is-probably-right-but-early/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-07-29-daniel-kokotajlo-is-probably-right-but-early/</guid><description>Daniel Kokotajlo&apos;s AI forecasts look extreme until you separate the mechanism from the calendar. The direction is plausible. The 2027-style timeline is the part I struggle to buy.</description><pubDate>Wed, 29 Jul 2026 11:00:00 GMT</pubDate></item><item><title>USA 2026, part 3: Manhattan, small legs, and zero spare hands</title><link>https://nikolajflojgaard.me/posts/2026/2026-07-26-usa-2026-part-3-new-york-city-with-kids/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-07-26-usa-2026-part-3-new-york-city-with-kids/</guid><description>The final leg of our 23 days in the States with Janice, Nelly, and Jamie: New York City on foot, big buildings doing free entertainment, and two adults trying to keep the moving parts attached.</description><pubDate>Sun, 26 Jul 2026 07:10:00 GMT</pubDate></item><item><title>USA 2026, part 2: big canyon, cold lake, warm island, small people</title><link>https://nikolajflojgaard.me/posts/2026/2026-07-26-usa-2026-part-2-grand-canyon-lake-tahoe-hawaii/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-07-26-usa-2026-part-2-grand-canyon-lake-tahoe-hawaii/</guid><description>The middle leg of our 23 days in the States with Janice, Nelly, and Jamie: the Grand Canyon being unfairly huge, Lake Tahoe changing the pace, San Francisco acting as a doorway, and Hawaii cashing the whole thing in.</description><pubDate>Sun, 26 Jul 2026 07:00:00 GMT</pubDate></item><item><title>USA 2026, part 1: jet lag, pool math, and Vegas at full volume</title><link>https://nikolajflojgaard.me/posts/2026/2026-07-26-usa-2026-part-1-palm-springs-las-vegas/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-07-26-usa-2026-part-1-palm-springs-las-vegas/</guid><description>The first leg of our 23 days in the States with Janice, Nelly, and Jamie: escaping LAX, letting Palm Springs repair us, and taking Las Vegas exactly as seriously as it deserves.</description><pubDate>Sun, 26 Jul 2026 06:50:00 GMT</pubDate></item><item><title>MCP needs an enterprise control plane</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-29-mcp-needs-an-enterprise-control-plane/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-29-mcp-needs-an-enterprise-control-plane/</guid><description>Model Context Protocol can make agents useful inside large companies, but only if it is wrapped in identity, policy, audit, ownership, and a serious gateway architecture.</description><pubDate>Mon, 29 Jun 2026 19:05:00 GMT</pubDate></item><item><title>Building a personal agent toolchain</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-25-building-a-personal-agent-toolchain/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-25-building-a-personal-agent-toolchain/</guid><description>A useful personal agent needs more than prompts. It needs a workbench, QA, memory discipline, durable task running, evaluation, and docs that stay honest in CI.</description><pubDate>Thu, 25 Jun 2026 09:57:00 GMT</pubDate></item><item><title>I came away from day 2 at DTW Ignite with a slightly annoying conclusion</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-24-i-came-away-from-day-2-at-dtw-ignite-with-a-slightly-annoying-conclusion/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-24-i-came-away-from-day-2-at-dtw-ignite-with-a-slightly-annoying-conclusion/</guid><description>Agents are the easy story. Changing the operating model is the hard one.</description><pubDate>Wed, 24 Jun 2026 14:19:09 GMT</pubDate></item><item><title>DTW Ignite day one: AI is everywhere, but telco still has to pay for it</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-23-dtw-ignite-day-one-ai-is-everywhere-but-telco-still-has-to-pay-for-it/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-23-dtw-ignite-day-one-ai-is-everywhere-but-telco-still-has-to-pay-for-it/</guid><description>First impressions from DTW Ignite 2026 in Copenhagen: agentic AI is all over the agenda, but telecom still has the awkward problem of making the economics work.</description><pubDate>Tue, 23 Jun 2026 14:30:00 GMT</pubDate></item><item><title>Agent ops beats agent demos</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-15-agent-ops-beats-agent-demos/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-15-agent-ops-beats-agent-demos/</guid><description>A useful personal AI agent is not one clever prompt. It is the boring operational loop around memory, recovery, context handoff, scheduled work, and cleanup.</description><pubDate>Mon, 15 Jun 2026 12:19:59 GMT</pubDate></item><item><title>FSD Supervised is finally real in Denmark</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-13-fsd-supervised-is-finally-real-in-denmark/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-13-fsd-supervised-is-finally-real-in-denmark/</guid><description>After more than 200 km with Tesla FSD Supervised in Denmark, the interesting part is not the approval headline. It is how quickly the feature turns from regulatory speculation into ordinary driving.</description><pubDate>Sat, 13 Jun 2026 16:20:00 GMT</pubDate></item><item><title>I Don&apos;t Want Investing to Be My Second Job</title><link>https://nikolajflojgaard.me/posts/2026/2026-06-08-i-dont-want-investing-to-be-my-second-job/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-06-08-i-dont-want-investing-to-be-my-second-job/</guid><description>Single-stock investing can become an unpaid second job that steals attention from work, family, and ordinary life. Sometimes the better portfolio decision is to make investing boring again.</description><pubDate>Mon, 08 Jun 2026 10:46:00 GMT</pubDate></item><item><title>Manhattan is ridiculously good with small children, and the playgrounds are everywhere</title><link>https://nikolajflojgaard.me/posts/2026/2026-05-31-manhattan-is-ridiculously-good-with-small-children-and-the-playgrounds-are-everywhere/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-05-31-manhattan-is-ridiculously-good-with-small-children-and-the-playgrounds-are-everywhere/</guid><description>A follow-up from our family babymoon: Manhattan turns out to be one of the easiest big cities to enjoy with a 1-year-old and a 5-year-old, mostly because the city gives you great walks, constant landmarks, and playground bailouts everywhere.</description><pubDate>Sun, 31 May 2026 15:45:00 GMT</pubDate></item><item><title>Why Great Companies Still Fail Gate 5</title><link>https://nikolajflojgaard.me/posts/2026/2026-05-17-why-great-companies-still-fail-gate-5/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-05-17-why-great-companies-still-fail-gate-5/</guid><description>A great business is not automatically a great stock. Gate 5 exists to stop me from paying for perfection and calling it discipline.</description><pubDate>Sun, 17 May 2026 10:10:00 GMT</pubDate></item><item><title>The 5 Gates I Use to Stop Myself Buying Good Stories and Bad Stocks</title><link>https://nikolajflojgaard.me/posts/2026/2026-05-17-the-5-gates-i-use-to-stop-myself-buying-good-stories-and-bad-stocks/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-05-17-the-5-gates-i-use-to-stop-myself-buying-good-stories-and-bad-stocks/</guid><description>Most stock frameworks are either too vague or too clever. This one is simple on purpose: five gates, a written template, and a hard rule that a great business is not automatically a great stock.</description><pubDate>Sun, 17 May 2026 10:05:00 GMT</pubDate></item><item><title>Helping Julian Build a Real School Project</title><link>https://nikolajflojgaard.me/posts/2026/2026-05-12-helping-julian-build-a-real-school-project/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-05-12-helping-julian-build-a-real-school-project/</guid><description>A fun, honest write-up of helping my wife’s younger brother build an Arduino motor-control school project, and why hands-on projects teach more than polished demos ever will.</description><pubDate>Tue, 12 May 2026 13:30:00 GMT</pubDate></item><item><title>The Agent Loop That Actually Works</title><link>https://nikolajflojgaard.me/posts/2026/2026-05-04-the-agent-loop-that-actually-works/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-05-04-the-agent-loop-that-actually-works/</guid><description>Most enterprise AI strategies are still too model-first. The real leverage comes from designing an agent loop that can survive contact with reality.</description><pubDate>Mon, 04 May 2026 11:58:00 GMT</pubDate></item><item><title>Babymoon in Manhattan, with kids, direct flight, and zero interest in rushing</title><link>https://nikolajflojgaard.me/posts/2026/2026-04-30-babymoon-in-manhattan-with-kids-direct-flight-and-zero-interest-in-rushing/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-04-30-babymoon-in-manhattan-with-kids-direct-flight-and-zero-interest-in-rushing/</guid><description>Why we are doing our Manhattan babymoon with a 1-year-old and a 5-year-old, flying direct to Newark, walking as much as possible, and refusing to turn New York into a checklist.</description><pubDate>Thu, 30 Apr 2026 13:35:00 GMT</pubDate></item><item><title>Home Brief turns Home Assistant daily briefs into something that actually persists</title><link>https://nikolajflojgaard.me/posts/2026/2026-04-30-home-brief-home-assistant-daily-briefs-that-persist/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-04-30-home-brief-home-assistant-daily-briefs-that-persist/</guid><description>A practical write-up of Home Brief, a Home Assistant integration that generates useful house summaries and supports a proper published daily brief instead of one-shot AI output.</description><pubDate>Thu, 30 Apr 2026 09:45:00 GMT</pubDate></item><item><title>Microsoft, Google, and Amazon are not just AI trades. They are the toll roads under the next compute cycle.</title><link>https://nikolajflojgaard.me/posts/2026/2026-04-28-hyperscalers-are-not-just-ai-trades-they-are-the-toll-roads-under-the-next-compute-cycle/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-04-28-hyperscalers-are-not-just-ai-trades-they-are-the-toll-roads-under-the-next-compute-cycle/</guid><description>A lot of the market still talks about hyperscalers like they are just expensive big tech again. I think that misses the point. If AI keeps scaling, Microsoft, Google, and Amazon are not side bets. They are the infrastructure layer the whole wave has to run through.</description><pubDate>Tue, 28 Apr 2026 11:05:00 GMT</pubDate></item><item><title>AI slop is everywhere now, and the internet already feels worse because of it</title><link>https://nikolajflojgaard.me/posts/2026/2026-04-23-ai-slop-is-everywhere-now/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-04-23-ai-slop-is-everywhere-now/</guid><description>AI-generated junk is flooding social media, YouTube, search, and comment sections faster than people can filter it. The problem is no longer whether AI can create content. It is whether the internet can stay worth using.</description><pubDate>Thu, 23 Apr 2026 15:05:00 GMT</pubDate></item><item><title>AI agents are not failing because they’re dumb. They’re failing because the system around them is sloppy.</title><link>https://nikolajflojgaard.me/posts/2026/2026-04-18-ai-agents-fail-because-the-system-is-sloppy/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-04-18-ai-agents-fail-because-the-system-is-sloppy/</guid><description>Most agent products do not break on model quality first. They break on messy context, brittle tools, fake evals, weak recovery paths, and sloppy release discipline.</description><pubDate>Sat, 18 Apr 2026 13:46:00 GMT</pubDate></item><item><title>Tesla FSD Supervised is finally approved in the Netherlands. Denmark may not be far behind.</title><link>https://nikolajflojgaard.me/posts/2026/2026-04-11-tesla-fsd-supervised-netherlands-denmark-timeline/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-04-11-tesla-fsd-supervised-netherlands-denmark-timeline/</guid><description>RDW has now approved Tesla FSD Supervised for use in the Netherlands. That does not put Denmark on the rollout list tomorrow, but it makes the path into the rest of Europe much more concrete.</description><pubDate>Sat, 11 Apr 2026 10:15:00 GMT</pubDate></item><item><title>We built a Household Chores system for Home Assistant that you can actually talk to</title><link>https://nikolajflojgaard.me/posts/2026/2026-03-31-household-chores-home-assistant-agent-tasks/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-03-31-household-chores-home-assistant-agent-tasks/</guid><description>A practical look at how I built Household Chores for Home Assistant, wired it into my assistant, and made task creation work through chat surfaces like iMessage and Telegram.</description><pubDate>Tue, 31 Mar 2026 08:00:00 GMT</pubDate></item><item><title>No one really understands how agentic AI will disrupt customer handling and every industry.</title><link>https://nikolajflojgaard.me/posts/2026/2026-03-27-no-one-really-understands-how-agentic-ai-will-disrupt-customer-handling-and-every-industry/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-03-27-no-one-really-understands-how-agentic-ai-will-disrupt-customer-handling-and-every-industry/</guid><description>Most people still think AI means a chatbot. The real shift is software that can act across your machine, tools, and workflows until the job is done.</description><pubDate>Fri, 27 Mar 2026 11:58:00 GMT</pubDate></item><item><title>Most enterprise AI still feels like letting a very confident intern loose in your systems.</title><link>https://nikolajflojgaard.me/posts/2026/2026-03-17-most-enterprise-ai-still-feels-like-letting-a-very-confident-intern-loose-in-your-systems/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-03-17-most-enterprise-ai-still-feels-like-letting-a-very-confident-intern-loose-in-your-systems/</guid><description>Most enterprise AI still feels like letting a very confident intern loose in your systems.</description><pubDate>Tue, 17 Mar 2026 06:41:21 GMT</pubDate></item><item><title>March 20 could be Tesla’s first real FSD Supervised opening in Europe</title><link>https://nikolajflojgaard.me/posts/2026/2026-03-16-rdw-elon-musk-march-20-fsd-supervised-europe/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-03-16-rdw-elon-musk-march-20-fsd-supervised-europe/</guid><description>Elon Musk says March 20 could bring FSD Supervised to Europe, likely through the Netherlands first. If that happens, RDW is the key gatekeeper — and Tesla’s current software rollout suggests this is more than a rename.</description><pubDate>Mon, 16 Mar 2026 13:55:00 GMT</pubDate></item><item><title>Sigenergy 5-in-1 bidirectional charging could help Denmarks grid problem</title><link>https://nikolajflojgaard.me/posts/2026/2026-03-08-sigenstor-bidirectional-charging-and-denmarks-grid/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-03-08-sigenstor-bidirectional-charging-and-denmarks-grid/</guid><description>Energinet&apos;s latest grid-capacity warning points to the real issue: Denmark does not just need more electricity. It needs more flexibility — and Sigenergy&apos;s 5-in-1 bidirectional setup is one way to think about that.</description><pubDate>Sun, 08 Mar 2026 19:20:00 GMT</pubDate></item><item><title>Anthropic and the Pentagon Dilemma</title><link>https://nikolajflojgaard.me/posts/2026/2026-03-08-anthropic-and-the-pentagon-dilemma/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-03-08-anthropic-and-the-pentagon-dilemma/</guid><description>When an AI company says it cares about safety and then walks toward defense work, the hard question isn&apos;t hypocrisy. It&apos;s whether responsible refusal is still possible once capability becomes geopolitical infrastructure.</description><pubDate>Sun, 08 Mar 2026 18:15:00 GMT</pubDate></item><item><title>FSD in Europe: The RDW Pathway and What It Means for Tesla&apos;s EU Rollout</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-23-fsd-europe-rdw-pathway/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-23-fsd-europe-rdw-pathway/</guid><description>RDW&apos;s recent certification activities suggest Tesla&apos;s Full Self-Driving could finally reach European roads. Here&apos;s what the regulatory pathway looks like and why integration architects should pay attention.</description><pubDate>Mon, 23 Feb 2026 08:00:00 GMT</pubDate></item><item><title>Nio&apos;s Battery Swap Milestone: 165,898 Swaps in One Day</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-22-nio-battery-swap-milestone/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-22-nio-battery-swap-milestone/</guid><description>Nio just proved battery swap technology can operate at true fuel-station scale. What does this mean for the future of EV charging?</description><pubDate>Sun, 22 Feb 2026 20:00:00 GMT</pubDate></item><item><title>AI speed without API governance is just faster technical debt.</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-19-ai-speed-without-api-governance-is-just-faster-technical-debt/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-19-ai-speed-without-api-governance-is-just-faster-technical-debt/</guid><description>AI speed without API governance is just faster technical debt.</description><pubDate>Thu, 19 Feb 2026 09:30:05 GMT</pubDate></item><item><title>On‑prem is the king again (but SaaS might not be dead yet).</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-17-on-prem-is-the-king-again-but-saas-might-not-be-dead-yet/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-17-on-prem-is-the-king-again-but-saas-might-not-be-dead-yet/</guid><description>On‑prem is the king again (but SaaS might not be dead yet).</description><pubDate>Tue, 17 Feb 2026 10:22:42 GMT</pubDate></item><item><title>Thinking of CRM without integration is like having a horse with no saddle.</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-16-thanks-to-thomas-fabricius-bjerre-mba-for-the-great-title-and-push-back-on-my-previous-pos/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-16-thanks-to-thomas-fabricius-bjerre-mba-for-the-great-title-and-push-back-on-my-previous-pos/</guid><description>In telco, the integration layer becomes the control plane for reliable cross-domain execution as agentic automation scales.</description><pubDate>Mon, 16 Feb 2026 10:50:21 GMT</pubDate></item><item><title>Playing around with Home Assistant, maybe this template is for you?</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-14-playing-around-with-home-assistant-maybe-this-template-is-for-you/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-14-playing-around-with-home-assistant-maybe-this-template-is-for-you/</guid><description>Playing around with Home Assistant, maybe this template is for you?</description><pubDate>Sat, 14 Feb 2026 12:13:14 GMT</pubDate></item><item><title>Agentic delivery for enterprise CRM: governance at the core</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-12-agentic-delivery-crm-governance/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-12-agentic-delivery-crm-governance/</guid><description>Agentic delivery (including OpenClaw-style orchestration) can speed up CRM innovation.</description><pubDate>Thu, 12 Feb 2026 08:13:22 GMT</pubDate></item><item><title>A common telecom transformation myth is: integration numbers and complexity should decline as maturity increases.</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-09-a-common-telecom-transformation-myth-is-integration-numbers-and-complexity-shoul/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-09-a-common-telecom-transformation-myth-is-integration-numbers-and-complexity-shoul/</guid><description>A common telecom transformation myth is: integration numbers and complexity should decline as maturity increases.</description><pubDate>Mon, 09 Feb 2026 08:15:29 GMT</pubDate></item><item><title>AI doesn’t replace architecture — it makes good architecture non‑negotiable.</title><link>https://nikolajflojgaard.me/posts/2026/2026-02-05-ai-doesn-t-replace-architecture-it-makes-good-architecture-non-negotiable/</link><guid isPermaLink="true">https://nikolajflojgaard.me/posts/2026/2026-02-05-ai-doesn-t-replace-architecture-it-makes-good-architecture-non-negotiable/</guid><description>AI doesn’t replace architecture — it makes good architecture non‑negotiable.</description><pubDate>Thu, 05 Feb 2026 09:48:18 GMT</pubDate></item></channel></rss>