<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Jakob Lange — Insights</title><description>Long-form articles, guides and field notes on AI engineering, sovereignty, compliance and the human side of security.</description><link>https://jakoblange.cloud/</link><language>en</language><atom:link href="https://jakoblange.cloud/rss.xml" rel="self" type="application/rss+xml"/><item><title>Determinism is built around the model, not into it</title><link>https://jakoblange.cloud/insights/determinism-is-built-around-the-model/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/determinism-is-built-around-the-model/</guid><description>A language model will not give the same answer twice, and no prompt changes that. What can be made deterministic are the properties that matter: what never happens, what always happens, in which order and shape, and what cannot ship unverified. A close reading of Claude Code shows where instructions end, where guarantees begin, and where people belong.</description><pubDate>Fri, 18 Sep 2026 00:00:00 GMT</pubDate><category>ai-agents</category><category>ai-engineering</category><category>context-engineering</category><category>ai-security</category><category>ai-act</category><author>hej@jakoblange.cloud</author></item><item><title>Build, buy or partner: an AI decision framework with real costs</title><link>https://jakoblange.cloud/insights/build-buy-or-partner-ai-decision-framework/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/build-buy-or-partner-ai-decision-framework/</guid><description>Every AI option buys three things at once: a capability, a dependency and an option on the future. A decision method with three gates, a three-year cost model that includes leaving, explicit lock-in and sovereignty scores — and why the honest answer is usually a hybrid.</description><pubDate>Mon, 07 Sep 2026 00:00:00 GMT</pubDate><category>ai-strategy</category><category>data-sovereignty</category><category>ai-engineering</category><category>platform</category><author>hej@jakoblange.cloud</author></item><item><title>Berlin&apos;s data leak: not paying the ransom is an architecture decision</title><link>https://jakoblange.cloud/insights/berlin-leak-not-paying-is-an-architecture-decision/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/berlin-leak-not-paying-is-an-architecture-decision/</guid><description>On 4 September 2026 the Rhysida group published 1.44 million files from two Berlin Senate administrations after the state refused a 30-bitcoin ransom. Refusing was right. What made it expensive was decided years earlier: identity, segmentation, data hygiene, egress monitoring and a consolidation that never happened.</description><pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate><category>cyber-resilience</category><category>infrastructure</category><category>security-behaviour</category><category>data-integrity</category><author>hej@jakoblange.cloud</author></item><item><title>Frontier AI has changed cyber security for good</title><link>https://jakoblange.cloud/insights/frontier-ai-changed-cyber-security-for-good/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/frontier-ai-changed-cyber-security-for-good/</guid><description>Frontier models turned expert attention — the scarce input on both sides of cyber security — into a metered commodity. What that changed for attackers and for defenders, why the exploit window collapsed to hours, and why &quot;nobody is looking at our custom code&quot; stopped being a defence.</description><pubDate>Thu, 03 Sep 2026 00:00:00 GMT</pubDate><category>ai-security</category><category>security-behaviour</category><category>platform</category><category>ai-strategy</category><author>hej@jakoblange.cloud</author></item><item><title>Which AI use cases pay for themselves within a year?</title><link>https://jakoblange.cloud/insights/which-ai-use-cases-pay-for-themselves/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/which-ai-use-cases-pay-for-themselves/</guid><description>A portfolio method for AI use cases: four value mechanisms and how much of each reaches the profit and loss, a business case built on measured baselines and an honest adoption assumption, a value–feasibility map with the AI Act drawn on it — and the business-development half most strategies forget.</description><pubDate>Thu, 03 Sep 2026 00:00:00 GMT</pubDate><category>ai-strategy</category><category>ai-act</category><category>knowledge-management</category><category>ai-engineering</category><author>hej@jakoblange.cloud</author></item><item><title>DORA for engineering teams: what the regulation actually asks of your systems</title><link>https://jakoblange.cloud/insights/dora-for-engineering-teams/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/dora-for-engineering-teams/</guid><description>A translation of the Digital Operational Resilience Act from legal text into engineering requirements — ICT risk, incident reporting, resilience testing, third-party risk — with what it means for teams that build and run AI and platform systems.</description><pubDate>Tue, 11 Aug 2026 00:00:00 GMT</pubDate><category>dora</category><category>data-integrity</category><category>infrastructure</category><category>ai-engineering</category><author>hej@jakoblange.cloud</author></item><item><title>Data sovereignty is an engineering property</title><link>https://jakoblange.cloud/insights/data-sovereignty-is-an-engineering-property/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/data-sovereignty-is-an-engineering-property/</guid><description>Sovereignty is decided in architecture, not in contracts: where data physically lives, which law can reach it, who holds the keys, and whether you can leave. A design method for EU-conform AI and platform hosting.</description><pubDate>Tue, 14 Jul 2026 00:00:00 GMT</pubDate><category>data-sovereignty</category><category>eu-hosting</category><category>deployment</category><category>infrastructure</category><author>hej@jakoblange.cloud</author></item><item><title>Context engineering: your token budget is your architecture</title><link>https://jakoblange.cloud/insights/context-engineering-token-budget-is-architecture/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/context-engineering-token-budget-is-architecture/</guid><description>Why most production failures of LLM systems are context failures, how tokenization shapes cost and quality, and a practical method for designing context windows as deliberately as you design a database schema.</description><pubDate>Tue, 16 Jun 2026 00:00:00 GMT</pubDate><category>context-engineering</category><category>ai-engineering</category><category>knowledge-management</category><author>hej@jakoblange.cloud</author></item><item><title>Social engineering in the age of generative AI: assess behaviour, not awareness</title><link>https://jakoblange.cloud/insights/social-engineering-tactics/</link><guid isPermaLink="true">https://jakoblange.cloud/insights/social-engineering-tactics/</guid><description>The classic social-engineering tactics, what generative AI has changed about them, why annual awareness training does not move the numbers, and how behaviour assessments and designed workflows do.</description><pubDate>Mon, 04 Mar 2024 00:00:00 GMT</pubDate><category>social-engineering</category><category>security-behaviour</category><category>research-methods</category><author>hej@jakoblange.cloud</author></item></channel></rss>