{"id":19990,"date":"2026-07-14T08:00:53","date_gmt":"2026-07-14T06:00:53","guid":{"rendered":"https:\/\/www.gtb.de\/?p=19990"},"modified":"2026-07-14T17:28:21","modified_gmt":"2026-07-14T15:28:21","slug":"when-the-agents-write-the-code-the-human-checks-themselves-2-2","status":"publish","type":"post","link":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/","title":{"rendered":"# When the Agents Write the Code, the Human Checks Themselves [2\/2]"},"content":{"rendered":"

## Part 2: The Price, or Three Cognitive Bottlenecks No Demo Ever Shows<\/h2><\/div>

*(A two-part series. Part 1 describes what a working agentic setup looks like. Part 2 shows the price the human pays for it and how to keep that price under control. Both parts can be read on their own.)*<\/p><\/div><\/div><\/div><\/div>

Picture a normal morning. The setup is warm, several agents run in parallel. One asks whether it may commit. The second has reported a red test and waits for a diagnosis. The third proposes an architecture switch. The fourth wants a decision on logging. Four decisions before the coffee goes cold. And that is the quiet part of the morning.<\/p>\n<\/div><\/div><\/div>

Agentic development here does not mean vibe coding, that is “the AI builds on the fly, I keep whatever sounds right”. It means an orchestrated setup of several specialized subagents, each with a clear mandate, running against a repository base with good test coverage, and a human who holds the strategic layer. How that setup looks in detail, why several models run side by side, and what makes it economically viable is covered in Part 1. This part is about the other half, the one that rarely gets told to the end: what lands on the human who supervises the agents.<\/p>\n

Most texts about agentic development sell speed. Ship faster, more output per day, less typing. All true. Except typing was never the exhausting part. Judging is. And that is exactly where three bottlenecks appear that I have come to take seriously.<\/p>\n<\/div><\/div><\/div><\/div>

\"J\u00f6rg<\/span><\/div>

J\u00f6rg Amelunxen<\/p>\n<\/div>

J\u00f6rg Amelunxen (M.Sc. Computer Science, University of Paderborn) is an AI Solutions Architect at mgm, with a path from lead developer through technical project lead to his current focus on AI- and agent-driven software development. In mgm’s cross-project AI & Data Engineering team he works on making agentic development production-ready in the enterprise: from multi-agent workflows and RAG systems, to MCP-based agent interfaces, to quality assurance in agent-driven processes. His project experience spans public sector, e-commerce, insurance, and finance. Most recently he designed the end-to-end architecture of a voice agent with its own intent classifier. He enjoys talking through these topics; if you would like to, you can find him easily on LinkedIn.<\/p>\n<\/div>

Ver\u00f6ffentlicht: 07.2026<\/p>\n<\/div><\/div><\/div>

## Bottleneck One: Decision Fatigue<\/h3><\/div>

In social psychology, decision fatigue has been described since Baumeister and colleagues in 1998 [1]<\/span>. The exact mechanism (ego depletion as a central reservoir of willpower) has been strongly called into question by a multi-lab preregistered replication with 23 labs and over 2000 participants [2]<\/span>. The phenomenon itself has not been: the quality of a decision measurably drops over the course of a sequence of decisions. The most robust evidence comes from a PNAS study of more than 1000 real parole decisions by eight experienced judges, observed over 50 court days [3]<\/span>. Right after a break or lunch, roughly 65 percent of applications are granted. Over the course of the following session, the rate falls gradually, in individual sessions down to near zero, and jumps back up after the next break. A conceptual review from health psychology confirms that the phenomenon shows up across professions [4]<\/span>. Clinical decision research provides a second, independent confirmation: a study of more than 21,000 outpatient visits in primary care practices shows that physicians increasingly prescribe antibiotics inappropriately over the course of each clinic session, peaking in the last hours before the lunch break [5]<\/span>. Same mechanism, different context, same direction.<\/p>\n

What happens in an agentic setup is the rapid translation of this mechanism into a developer’s day. Where I used to make maybe eight conscious decisions in two hours (“do I take this library, do I refactor this method, do I write this test”), it now feels like forty. In the same amount of time, the agents produce a multiple of the material that needs deciding. Typing was never the exhausting part. Judging is.<\/p>\n

For testers this is doubly interesting. Anyone who in the classic setup already held the role of “reviewer of code someone else wrote” knows this fatigue already. In the agentic world it is not reinvented but generalized: suddenly everyone working with agents is constantly in the reviewer position. What used to be a time-boxed task in classic code review becomes a permanent state.<\/p>\n

The everyday effect sounds harmless. You start clicking “OK” faster. You skim diffs you should really be reading line by line. You accept architecture proposals where yesterday you would still have asked “why, actually”. Nobody says it out loud. But almost everyone who works honestly with the setup knows that moment in the late afternoon.<\/p>\n<\/div>

\"Decision<\/span><\/div>
<\/div><\/div>

## Bottleneck Two: Context Switching<\/h3><\/div>

The second cost factor is context switching. The cost of context switches has been folklore in software engineering practice for decades and is well documented in CHI research. A field study of knowledge workers finds an average of roughly 25 minutes until an interrupted task is resumed, typically after more than two other tasks in between [6]<\/span>. A follow-up study shows that interrupted tasks do get finished measurably faster, but at a measurable price of more stress, frustration, and effort [7]<\/span>. For programming work specifically, the picture is even clearer. Parnin & Rugaber in 2011 empirically examine how developers find their way back into the code after an interruption, and find that they do not simply pick up where they left off; they visibly invest time in recovery strategies such as code notes, deliberately left compile errors as bookmarks, or rereading the last commit, because the mental stack is deeper than in generic knowledge work [8]<\/span>.<\/p>\n

In agentic supervision, the context switch does not sit between the hours but inside the block of work. You jump from the refactor review to the E2E diagnosis to the architecture decision to the logging detail. Four different contexts, four different mental models, all in thirty minutes. Every jump costs setup time. And because the agents work in parallel and do not wait, the temptation is strong to take each one as it comes in.<\/p>\n

What helps is classic engineering discipline in new clothes: class-batched reviews. All refactor reviews in one block. All architecture decisions in a second. Test diagnoses in a third. This sounds obvious and is not; the agents behave like an open-plan office where someone is constantly standing at your desk.<\/p>\n

The trick is not to run more agents at once. The trick is that only one kind of question reaches you at a time.<\/p>\n<\/div>

<\/div><\/div>

## Bottleneck Three: Attention Span<\/h3><\/div>

The third bottleneck is the most insidious, because it does not feel like fatigue but like routine. Attention span under continuous review.<\/p>\n

Cognitive psychology knows the phenomenon as the vigilance decrement, documented since Mackworth’s classic study of radar monitoring in the Second World War [9]<\/span>: someone who has to attentively monitor something over a long time, where an error appears only rarely, loses the ability to detect it. In the lab, Mackworth found a 10 to 15 percent drop in detection rate within the first thirty minutes, with further deterioration afterward. A much-cited review from human factors carries the pattern over to modern automation: when a human monitors an automated pipeline while also having their own tasks, complacency rises, and simple practice does not help against it [10]<\/span>. More recent research specifically on human oversight of AI suggestions shows the pattern in its present-day form. Vasconcelos and colleagues demonstrate empirically in 2023 that users tend to accept AI suggestions without checking them themselves as soon as they perceive the cognitive effort of verification as too high; even well-constructed explanations only partly reduce this over-reliance, because the human runs a simple cost-benefit calculation [11]<\/span>. In agentic supervision this is exactly the situation: many suggestions, each individually plausible, real verification more expensive than acceptance.<\/p>\n

In practice this means: after thirty minutes of continuous review work, the detection rate drops. You stop seeing bugs you would have caught immediately when fresh. And because the agents keep producing, this effect adds up fast.<\/p>\n

From my own practice, a single observation, not a study finding: when I check agent output for longer than half an hour at a stretch, I find things on rereading after a break that I did not see on the first pass. This is the same mechanism that code review research has known since the beginnings of formal inspections; as early as 1976, Fagan recommends limiting an inspection session to at most two hours and reviewing no more than 150 lines per hour, because otherwise the defect detection rate drops significantly [12]<\/span>. A modern empirical study of code review practice at Microsoft extends the picture: in today’s asynchronous practice, reviews revolve less around pure defect detection than expected; a large share of the documented reviewer comments aims at code understanding and knowledge transfer within the team [13]<\/span>. That does not make the attention demand on the reviewer smaller, it shifts it. In the agentic world the same effect just becomes permanent instead of situational.<\/p>\n<\/div>

<\/div><\/div>

## What the Setup Needs Against This<\/h3><\/div>

When the three bottlenecks come together, they form a pattern I have come to take seriously. Three levers that work in my daily practice:<\/p>\n

First, a decision-budget logic. A long-term study of 27 CEOs of large companies, tracked around the clock over 13 weeks, shows the pattern cleanly: top executives spend the largest part of their working time in meetings, where the main activities are prioritizing, sorting, and handing off; the real strategic decisions of their own are a clear minority [14]<\/span>. In an agentic setup, by contrast, you first decide everything yourself. You have to correct that explicitly: which decisions are reversible and cheap (a code comment, a small refactor in an internal library) and pass without review? Which are reversible and expensive (a larger refactor, an architecture detail)? Which are not reversible (a schema change in production, an external interface)? The first category moves to auto-approve, the third stays strictly with the human.<\/p>\n

Second, lead in groups instead of darting between individual agents. Three fixed slots a day are the clean ideal and in practice rarely feasible; in a warm agent fleet you have to hook in somewhere all the time. What does work, once you are no longer running just four agents in parallel but rather twelve or sixteen, is bundling them into thematic groups with their own focus. One group works on the backend, one on the tests, one on documentation and small bug fixes. You actively devote yourself to one group at a time, instead of jumping between the agents of different groups.<\/p>\n

This works like a dog walker with several groups of dogs at once. He actively leads one group; the others stand around, look about, or run toward the goal where they are already playing anyway. When he switches, he switches at the group level, not between individual dogs of two parallel groups. That is exactly the difference between functioning parallel supervision and chronic darting about.<\/p>\n

Third, a thirty-minute cap on continuous review work. With a break afterward that is not a screen again. Movement, coffee, staring at a wall. The vigilance research is clear enough that you do not need to optimize further.<\/p>\n<\/div>

<\/div><\/div>

## The Question That Counts in the End<\/h3><\/div>

Anyone who measures agentic development only along the axis of “faster and cheaper” is measuring the wrong thing. Speed has become a given, not a differentiator. What sets software apart today is no longer that it gets shipped, but that it is unique. That it does something that is not also sitting in every second repository. That the human who is responsible for it is genuinely still in the loop, with the three bottlenecks above under control.<\/p>\n

For the testing community, I think, this means two things.<\/p>\n

One: the role of the tester does not disappear in an agentic world. It becomes more central. What the agents produce has to be checked by someone who holds the context. That is exactly the skill good testers have always had. It is just now being called upon under new conditions.<\/p>\n

The other: the three bottlenecks above apply especially to people in review roles. Whoever takes the setup seriously builds guardrails. Whoever does not runs the risk of scrapping the AI in half a year with the argument, “it was not for us”. The difference does not lie in the AI. It lies in the architecture around it.<\/p>\n<\/div>

<\/div><\/div><\/div><\/div>
<\/div><\/div>

## References<\/h3><\/div>
\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
[1]<\/td>\nBaumeister, R.F., Bratslavsky, E., Muraven, M., & Tice, D.M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74(5), 1252-1265. https:\/\/doi.org\/10.1037\/0022-3514.74.5.1252<\/a><\/td>\n<\/tr>\n
[2]<\/td>\nHagger, M.S., Chatzisarantis, N.L.D., Alberts, H., et al. (2016). A multilab preregistered replication of the ego-depletion effect. Perspectives on Psychological Science, 11(4), 546-573. https:\/\/doi.org\/10.1177\/1745691616652873<\/a><\/td>\n<\/tr>\n
[3]<\/td>\nDanziger, S., Levav, J., & Avnaim-Pesso, L. (2011). Extraneous factors in judicial decisions. Proceedings of the National Academy of Sciences, 108(17), 6889-6892. https:\/\/doi.org\/10.1073\/pnas.1018033108<\/a><\/td>\n<\/tr>\n
[4]<\/td>\nPignatiello, G.A., Martin, R.J., & Hickman, R.L. (2020). Decision fatigue: A conceptual analysis. Journal of Health Psychology, 25(1), 123-135. https:\/\/doi.org\/10.1177\/1359105318763510<\/a><\/td>\n<\/tr>\n
[5]<\/td>\nLinder, J.A., Doctor, J.N., Friedberg, M.W., Reyes Nieva, H., Birks, C., Meeker, D., & Fox, C.R. (2014). Time of day and the decision to prescribe antibiotics. JAMA Internal Medicine, 174(12), 2029-2031. https:\/\/doi.org\/10.1001\/jamainternmed.2014.5225<\/a><\/td>\n<\/tr>\n
[6]<\/td>\nMark, G., Gonzalez, V.M., & Harris, J. (2005). No task left behind? Examining the nature of fragmented work. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’05), 321-330. https:\/\/doi.org\/10.1145\/1054972.1055017<\/a><\/td>\n<\/tr>\n
[7]<\/td>\nMark, G., Gudith, D., & Klocke, U. (2008). The cost of interrupted work: more speed and stress. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’08), 107-110. https:\/\/doi.org\/10.1145\/1357054.1357072<\/a><\/td>\n<\/tr>\n
[8]<\/td>\nParnin, C., & Rugaber, S. (2011). Resumption strategies for interrupted programming tasks. Software Quality Journal, 19(1), 5-34. https:\/\/doi.org\/10.1007\/s11219-010-9104-9<\/a><\/td>\n<\/tr>\n
[9]<\/td>\nMackworth, N.H. (1948). The breakdown of vigilance during prolonged visual search. Quarterly Journal of Experimental Psychology, 1(1), 6-21. https:\/\/doi.org\/10.1080\/17470214808416738<\/a><\/td>\n<\/tr>\n
[10]<\/td>\nParasuraman, R., & Manzey, D.H. (2010). Complacency and bias in human use of automation: An attentional integration. Human Factors, 52(3), 381-410. https:\/\/doi.org\/10.1177\/0018720810376055<\/a><\/td>\n<\/tr>\n
[11]<\/td>\nVasconcelos, H., J\u00f6rke, M., Grunde-McLaughlin, M., Gerstenberg, T., Bernstein, M.S., & Krishna, R. (2023). Explanations can reduce overreliance on AI systems during decision-making. Proceedings of the ACM on Human-Computer Interaction, 7(CSCW1), 1-38. https:\/\/doi.org\/10.1145\/3579605<\/a><\/td>\n<\/tr>\n
[12]<\/td>\nFagan, M.E. (1976). Design and code inspections to reduce errors in program development. IBM Systems Journal, 15(3), 182-211. https:\/\/doi.org\/10.1147\/sj.153.0182<\/a><\/td>\n<\/tr>\n
[13]<\/td>\nBacchelli, A., & Bird, C. (2013). Expectations, outcomes, and challenges of modern code review. In Proceedings of the 2013 International Conference on Software Engineering (ICSE ’13), 712-721. https:\/\/doi.org\/10.1109\/ICSE.2013.6606617<\/a><\/td>\n<\/tr>\n
[14]<\/td>\nPorter, M.E., & Nohria, N. (2018). How CEOs manage time. Harvard Business Review, 96(4), 42-51. https:\/\/hbr.org\/2018\/07\/how-ceos-manage-time<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div><\/div><\/div><\/div><\/div><\/div><\/div><\/div>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":6,"featured_media":19989,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[81],"tags":[91,84],"class_list":["post-19990","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-blog","tag-quality-engineering","tag-test"],"acf":[],"yoast_head":"\n# When the Agents Write the Code, the Human Checks Themselves [2\/2] - German Testing Board<\/title>\n<meta name=\"description\" content=\"Decision fatigue, context switching, and fading attention: three cognitive bottlenecks of supervising agents, with evidence and concrete countermeasures.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"# When the Agents Write the Code, the Human Checks Themselves [2\/2] - German Testing Board\" \/>\n<meta property=\"og:description\" content=\"Decision fatigue, context switching, and fading attention: three cognitive bottlenecks of supervising agents, with evidence and concrete countermeasures.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/\" \/>\n<meta property=\"og:site_name\" content=\"German Testing Board\" \/>\n<meta property=\"article:published_time\" content=\"2026-07-14T06:00:53+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-07-14T15:28:21+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.gtb.de\/wp-content\/uploads\/2026\/07\/Heroimage-Post08-1024x462.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"462\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Dr. Armin Metzger\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@GTB_ISTQB\" \/>\n<meta name=\"twitter:site\" content=\"@GTB_ISTQB\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Dr. Armin Metzger\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"12 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/\"},\"author\":{\"name\":\"Dr. Armin Metzger\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#\\\/schema\\\/person\\\/580154dcda34782d68a1503d906ba00a\"},\"headline\":\"# When the Agents Write the Code, the Human Checks Themselves [2\\\/2]\",\"datePublished\":\"2026-07-14T06:00:53+00:00\",\"dateModified\":\"2026-07-14T15:28:21+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/\"},\"wordCount\":12411,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.gtb.de\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Heroimage-Post08.png\",\"keywords\":[\"Quality Engineering\",\"Test\"],\"articleSection\":[\"Blog\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/\",\"url\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/\",\"name\":\"# When the Agents Write the Code, the Human Checks Themselves [2\\\/2] - German Testing Board\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.gtb.de\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Heroimage-Post08.png\",\"datePublished\":\"2026-07-14T06:00:53+00:00\",\"dateModified\":\"2026-07-14T15:28:21+00:00\",\"description\":\"Decision fatigue, context switching, and fading attention: three cognitive bottlenecks of supervising agents, with evidence and concrete countermeasures.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#primaryimage\",\"url\":\"https:\\\/\\\/www.gtb.de\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Heroimage-Post08.png\",\"contentUrl\":\"https:\\\/\\\/www.gtb.de\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Heroimage-Post08.png\",\"width\":2392,\"height\":1080,\"caption\":\"Wenn die Agenten den Code schreiben, pr\u00fcft der Mensch sich selbst\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/blog\\\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"# When the Agents Write the Code, the Human Checks Themselves [2\\\/2]\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#website\",\"url\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/\",\"name\":\"German Testing Board\",\"description\":\"Software.Testing.Excellence\",\"publisher\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#organization\",\"name\":\"German Testing Board e. V.\",\"url\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/www.gtb.de\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/gtb-logo.png\",\"contentUrl\":\"https:\\\/\\\/www.gtb.de\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/gtb-logo.png\",\"width\":224,\"height\":183,\"caption\":\"German Testing Board e. V.\"},\"image\":{\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/GTB_ISTQB\",\"https:\\\/\\\/de.linkedin.com\\\/company\\\/german-testing-board\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/www.gtb.de\\\/en\\\/#\\\/schema\\\/person\\\/580154dcda34782d68a1503d906ba00a\",\"name\":\"Dr. Armin Metzger\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/70d5768f24f60d4aa0012915fb37c2937223f49270944aa45222f8a20583d028?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/70d5768f24f60d4aa0012915fb37c2937223f49270944aa45222f8a20583d028?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/70d5768f24f60d4aa0012915fb37c2937223f49270944aa45222f8a20583d028?s=96&d=mm&r=g\",\"caption\":\"Dr. Armin Metzger\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"# When the Agents Write the Code, the Human Checks Themselves [2\/2] - German Testing Board","description":"Decision fatigue, context switching, and fading attention: three cognitive bottlenecks of supervising agents, with evidence and concrete countermeasures.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/","og_locale":"en_US","og_type":"article","og_title":"# When the Agents Write the Code, the Human Checks Themselves [2\/2] - German Testing Board","og_description":"Decision fatigue, context switching, and fading attention: three cognitive bottlenecks of supervising agents, with evidence and concrete countermeasures.","og_url":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/","og_site_name":"German Testing Board","article_published_time":"2026-07-14T06:00:53+00:00","article_modified_time":"2026-07-14T15:28:21+00:00","og_image":[{"width":1024,"height":462,"url":"https:\/\/www.gtb.de\/wp-content\/uploads\/2026\/07\/Heroimage-Post08-1024x462.png","type":"image\/png"}],"author":"Dr. Armin Metzger","twitter_card":"summary_large_image","twitter_creator":"@GTB_ISTQB","twitter_site":"@GTB_ISTQB","twitter_misc":{"Written by":"Dr. Armin Metzger","Est. reading time":"12 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#article","isPartOf":{"@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/"},"author":{"name":"Dr. Armin Metzger","@id":"https:\/\/www.gtb.de\/en\/#\/schema\/person\/580154dcda34782d68a1503d906ba00a"},"headline":"# When the Agents Write the Code, the Human Checks Themselves [2\/2]","datePublished":"2026-07-14T06:00:53+00:00","dateModified":"2026-07-14T15:28:21+00:00","mainEntityOfPage":{"@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/"},"wordCount":12411,"commentCount":0,"publisher":{"@id":"https:\/\/www.gtb.de\/en\/#organization"},"image":{"@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#primaryimage"},"thumbnailUrl":"https:\/\/www.gtb.de\/wp-content\/uploads\/2026\/07\/Heroimage-Post08.png","keywords":["Quality Engineering","Test"],"articleSection":["Blog"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/","url":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/","name":"# When the Agents Write the Code, the Human Checks Themselves [2\/2] - German Testing Board","isPartOf":{"@id":"https:\/\/www.gtb.de\/en\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#primaryimage"},"image":{"@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#primaryimage"},"thumbnailUrl":"https:\/\/www.gtb.de\/wp-content\/uploads\/2026\/07\/Heroimage-Post08.png","datePublished":"2026-07-14T06:00:53+00:00","dateModified":"2026-07-14T15:28:21+00:00","description":"Decision fatigue, context switching, and fading attention: three cognitive bottlenecks of supervising agents, with evidence and concrete countermeasures.","breadcrumb":{"@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#primaryimage","url":"https:\/\/www.gtb.de\/wp-content\/uploads\/2026\/07\/Heroimage-Post08.png","contentUrl":"https:\/\/www.gtb.de\/wp-content\/uploads\/2026\/07\/Heroimage-Post08.png","width":2392,"height":1080,"caption":"Wenn die Agenten den Code schreiben, pr\u00fcft der Mensch sich selbst"},{"@type":"BreadcrumbList","@id":"https:\/\/www.gtb.de\/en\/blog\/when-the-agents-write-the-code-the-human-checks-themselves-2-2\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.gtb.de\/en\/"},{"@type":"ListItem","position":2,"name":"# When the Agents Write the Code, the Human Checks Themselves [2\/2]"}]},{"@type":"WebSite","@id":"https:\/\/www.gtb.de\/en\/#website","url":"https:\/\/www.gtb.de\/en\/","name":"German Testing Board","description":"Software.Testing.Excellence","publisher":{"@id":"https:\/\/www.gtb.de\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.gtb.de\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.gtb.de\/en\/#organization","name":"German Testing Board e. V.","url":"https:\/\/www.gtb.de\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.gtb.de\/en\/#\/schema\/logo\/image\/","url":"https:\/\/www.gtb.de\/wp-content\/uploads\/2023\/10\/gtb-logo.png","contentUrl":"https:\/\/www.gtb.de\/wp-content\/uploads\/2023\/10\/gtb-logo.png","width":224,"height":183,"caption":"German Testing Board e. V."},"image":{"@id":"https:\/\/www.gtb.de\/en\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/GTB_ISTQB","https:\/\/de.linkedin.com\/company\/german-testing-board"]},{"@type":"Person","@id":"https:\/\/www.gtb.de\/en\/#\/schema\/person\/580154dcda34782d68a1503d906ba00a","name":"Dr. Armin Metzger","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/70d5768f24f60d4aa0012915fb37c2937223f49270944aa45222f8a20583d028?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/70d5768f24f60d4aa0012915fb37c2937223f49270944aa45222f8a20583d028?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/70d5768f24f60d4aa0012915fb37c2937223f49270944aa45222f8a20583d028?s=96&d=mm&r=g","caption":"Dr. Armin Metzger"}}]}},"_links":{"self":[{"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/posts\/19990","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/comments?post=19990"}],"version-history":[{"count":8,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/posts\/19990\/revisions"}],"predecessor-version":[{"id":20069,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/posts\/19990\/revisions\/20069"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/media\/19989"}],"wp:attachment":[{"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/media?parent=19990"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/categories?post=19990"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.gtb.de\/en\/wp-json\/wp\/v2\/tags?post=19990"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}