<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>Ivan Pelivanovic, Author at ShiftMag</title>
	<atom:link href="https://shiftmag.dev/author/ivan-pelivanovic/feed/" rel="self" type="application/rss+xml" />
	<link>https://shiftmag.dev/author/ivan-pelivanovic/</link>
	<description>Insightful engineering content &#38; community</description>
	<lastBuildDate>Tue, 10 Nov 2026 15:00:34 +0000</lastBuildDate>
	<language>en-GB</language>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.1.4</generator>

<image>
	<url>https://shiftmag.dev/wp-content/uploads/2026/09/cropped-Shift-32x32.webp</url>
	<title>Ivan Pelivanovic, Author at ShiftMag</title>
	<link>https://shiftmag.dev/author/ivan-pelivanovic/</link>
	<width>32</width>
	<height>32</height>
</image> 
	<item>
		<title>AI Agents Shouldn’t Set Their Own Security Boundaries</title>
		<link>https://shiftmag.dev/ai-agents-shouldnt-set-their-own-security-boundaries-12827/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Tue, 10 Nov 2026 14:30:11 +0000</pubDate>
				<category><![CDATA[Artificial Intelligence]]></category>
		<category><![CDATA[Security]]></category>
		<category><![CDATA[AI agents]]></category>
		<category><![CDATA[AI security]]></category>
		<category><![CDATA[NVIDIA]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=12827</guid>

					<description><![CDATA[<p>"They just see friction versus no friction." That is how Ali Golshan, Senior Director of AI Software at NVIDIA, describes AI agents facing security rules.</p>
<p>The post <a href="https://shiftmag.dev/ai-agents-shouldnt-set-their-own-security-boundaries-12827/">AI Agents Shouldn’t Set Their Own Security Boundaries</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img fetchpriority="high" decoding="async" width="1200" height="720" src="https://shiftmag.dev/wp-content/uploads/2026/10/Ali.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/10/Ali.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/10/Ali-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/10/Ali-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/10/Ali-768x461.jpg 768w" sizes="(max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph">AI agents use tools, reach into internal systems and get work done with barely any human involved. New security problems turn up almost weekly. What if an agent&#8217;s own instructions block it, and it finds a way around them just to finish the job?</p>



<p class="wp-block-paragraph">At <a href="https://www.nvidia.com/en-eu/gtc/" target="_blank" rel="noreferrer noopener">NVIDIA&#8217;s GTC Berlin</a> I was at a session where <strong>Ali Golshan</strong>, Senior Director of AI Software at NVIDIA presented an approach that treats agent security as a problem spanning the entire software stack, from the application and runtime to the underlying infrastructure.</p>



<p class="wp-block-paragraph">His central argument was that<strong> security cannot depend solely on whether a model follows instructions</strong>.</p>



<h2 class="wp-block-heading" id="enforce-rules-where-agents-cant-reach">Enforce rules where agents can&#8217;t reach</h2>



<p class="wp-block-paragraph">One big issue Golshan pointed out is that <strong>agents don’t understand restrictions the way developers meant them to</strong>. If an agent can’t get to a tool or resource it might look for another way to get the job done. In systems this can lead the agent to drift from its original goal or even find a hidden path to sensitive data. As he said:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">They just see friction versus no friction.</p>
</blockquote>



<p class="wp-block-paragraph">The point is that an agent <strong>may treat a security restriction as an obstacle to overcome</strong> rather than a boundary it must respect. If controls exist only in the application layer, where the agent can reason about them, they may not provide a sufficiently reliable enforcement mechanism.</p>



<p class="wp-block-paragraph">Ali argued that security controls should therefore be moved into layers that <strong>operate independently of the agent</strong>. The agent can propose an action, but a separate mechanism should determine whether that action is permitted and enforce the decision. This means separating an agent’s ability to plan from its authority to act.</p>



<h2 class="wp-block-heading" id="every-agent-can-follow-the-rules-while-the-system-as-a-whole-breaks-them">Every agent can follow the rules while the system as a whole breaks them</h2>



<p class="wp-block-paragraph">The problem becomes more complicated when multiple agents collaborate. Golshan described a scenario in which one agent can access GitHub but cannot access the public internet, while another can publish content online but has no access to GitHub.</p>



<p class="wp-block-paragraph">Each agent operates within its individual permissions. Yet if the first agent delegates work to the second, or passes information between them, their combined capabilities may exceed the limits an organisation intended to enforce:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">However, you have three agents that their collective capabilities now exceed your global policy. How do you verify these unintended exfiltration paths?</p>
</blockquote>



<p class="wp-block-paragraph">This illustrates why <strong>checking permissions for individual agents may not be enough</strong>. In multi-agent systems, security controls must also account for delegation and the combined effects of actions across an entire workflow.</p>



<p class="wp-block-paragraph"><strong>Longrunning agents </strong>introduce another challenge. As they work, call tools or delegate tasks, their access requirements can change. Persistent credentials and permissions may accumulate unnecessary risk if they remain active after a task is complete.</p>



<p class="wp-block-paragraph">A more restrictive approach is to grant agents only the permissions they need, for as long as they need them, and revoke access when the relevant work is finished.</p>



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/10/AliG-1-1024x614.jpg?x32039" alt="" class="wp-image-12989" srcset="https://shiftmag.dev/wp-content/uploads/2026/10/AliG-1-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/10/AliG-1-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/10/AliG-1-767x460.jpg 767w, https://shiftmag.dev/wp-content/uploads/2026/10/AliG-1.jpg 1200w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Credit: Ivan Pelivanović</figcaption></figure>



<p class="wp-block-paragraph"><strong>The risks also grow as agents operate for longer periods</strong>. Ali pointed to long-running agents and multi-agent fleets as two developments that make security harder to manage. Over time, an agent may accumulate access, credentials and session state, while collaboration introduces risks that cannot be assessed by looking at each agent in isolation.</p>



<h2 class="wp-block-heading" id="how-to-solve-this">How to solve this?</h2>



<p class="wp-block-paragraph">He advocated a layered approach that combines isolation, narrowly scoped permissions, monitoring and the ability to revoke access quickly.</p>



<p class="wp-block-paragraph">The principle is to <strong>reduce the damage an agent can cause if it makes a mistake</strong>, is manipulated by malicious input or exploits a vulnerability in its environment. Sensitive credentials should remain outside the agent’s direct reach wherever possible, while access to external systems should pass through independently enforced controls.</p>



<p class="wp-block-paragraph">NVIDIA presented its own implementation of this approach through OpenShell, an agent runtime, alongside infrastructure-level controls associated with its BlueField platform and Sentry security framework.</p>



<p class="wp-block-paragraph">The proposed architecture aims to move some enforcement beyond the software environment in which an agent operates. The broader idea is that an agent should not be able to change or bypass the security controls governing its own execution.</p>



<p class="wp-block-paragraph">Ali described this as a <strong>progression from application-level controls to runtime isolation</strong> and, ultimately, hardware-level enforcement. The reasoning is that a runtime is still software and may itself contain vulnerabilities. NVIDIA’s proposed architecture aims to place some security functions in a separate infrastructure layer that the agent cannot modify or bypass directly.</p>



<h2 class="wp-block-heading" id="agent-security-cant-be-solved-one-vendor-at-a-time">Agent security can&#8217;t be solved one vendor at a time</h2>



<p class="wp-block-paragraph">Enterprise environments <strong>combine different models</strong>, cloud platforms, tools and internal systems. If each provider implements its own incompatible approach to identity, permissions and delegation, organisations may struggle to enforce consistent controls across their infrastructure. Shared approaches to agent identity, authorisation, delegation and auditability could let them apply one security policy across different systems. NVIDIA pointed to the Open Secure AI Alliance as work toward open standards for agentic systems.</p>



<p class="wp-block-paragraph">Before deploying agents, ask what each one can access, what it can change, and whether those limits still hold when several agents work together.</p>
<p>The post <a href="https://shiftmag.dev/ai-agents-shouldnt-set-their-own-security-boundaries-12827/">AI Agents Shouldn’t Set Their Own Security Boundaries</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Jensen Huang&#8217;s Advice to Young Developers: &#8220;Become the World&#8217;s Best User of AI&#8221;</title>
		<link>https://shiftmag.dev/jensen-huangs-advice-to-young-developers-become-the-worlds-best-user-of-ai-13274/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Tue, 03 Nov 2026 14:10:23 +0000</pubDate>
				<category><![CDATA[Artificial Intelligence]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[Jensen Huang]]></category>
		<category><![CDATA[NVIDIA]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=13274</guid>

					<description><![CDATA[<p>Nvidia's CEO says young people shouldn't fear AI taking their jobs. They should learn to use it to do things they couldn't do before.</p>
<p>The post <a href="https://shiftmag.dev/jensen-huangs-advice-to-young-developers-become-the-worlds-best-user-of-ai-13274/">Jensen Huang&#8217;s Advice to Young Developers: &#8220;Become the World&#8217;s Best User of AI&#8221;</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img decoding="async" width="1200" height="720" src="https://shiftmag.dev/wp-content/uploads/2026/11/gtc-berlin.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/11/gtc-berlin.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/11/gtc-berlin-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/11/gtc-berlin-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/11/gtc-berlin-768x461.jpg 768w" sizes="(max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph"></p>



<p class="wp-block-paragraph">At Berlin’s packed Tempodrom, NVIDIA founder and CEO <strong>Jensen Huang</strong> was welcomed like a rock star at the company’s <a href="https://www.nvidia.com/en-eu/gtc/" target="_blank" rel="noreferrer noopener">GTC conference</a> where he shared the <a href="https://www.nvidia.com/en-eu/gtc/keynote/" target="_blank" rel="noreferrer noopener">big announcements about the future of AI</a> and NVIDIA&#8217;s products.</p>



<p class="wp-block-paragraph">The following day, Huang sat down with journalists, analysts and content creators for a dedicated Q&amp;A session. Alongside around 70 other attendees, we had the opportunity to hear what the head of <a href="https://companiesmarketcap.com/nvidia/marketcap/" target="_blank" rel="noreferrer noopener">the world’s most valuable company</a> thinks about whether AI will replace junior roles and  whether it is actually making us more productive.</p>



<p class="wp-block-paragraph">And no, that’s not AI photo.<strong> I really did get a photo with Jensen</strong>!</p>



<h2 class="wp-block-heading" id="with-ai-we-can-do-things-we-couldnt-before">With AI, we can do things we couldn&#8217;t before</h2>



<p class="wp-block-paragraph">The conversation about artificial intelligence at work often comes down to how quickly it can generate code, summarise a document or write a piece of content. Huang believes <strong>we should be thinking about productivity in broader terms</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">There are two ways to think about productivity. One is, of course, that whatever you used to do, you can now do more efficiently. But the better way of thinking about AI is that you can now do things you couldn&#8217;t do before.</p>
</blockquote>



<p class="wp-block-paragraph">As an example, he pointed to engineering and design. Rather than manually working through a limited number of possible solutions, engineers can use AI to explore a much wider design space while accounting for constraints such as cost, energy efficiency and manufacturability. The value of AI, in this case is that <strong>it opens up possibilities that would have been difficult to explore otherwise</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">AI can give you thousands of suggestions. That is real productivity, and this principle can be applied to almost any industry.</p>
</blockquote>



<p class="wp-block-paragraph">Huang believes <strong>we should stop measuring productivity solely by how much faster we can complete existing tasks</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">That is just not very ambitious. We should raise our ambitions, not just give ourselves more free time. I think we have always thought about productivity in precisely the wrong way. Technological progress works in the opposite way to what people often expect. Every time we invented a new technology, the economy got bigger and, frankly, we got busier.</p>
</blockquote>



<h2 class="wp-block-heading" id="what-about-juniors">What about juniors?</h2>



<p class="wp-block-paragraph">Huang was also asked about the global jobs market and the<strong> lack of entry-level opportunities</strong> for people graduating from university or just starting their careers:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Of course, some jobs will become less necessary than before because of AI, or perhaps less interesting. But new jobs, new companies and new businesses will emerge. That is the nature of the world. These changes are inevitable, and it is a good thing that the world keeps changing.</p>
</blockquote>



<p class="wp-block-paragraph">So how should young people prepare for those changes? We asked Huang what he would do if he were starting his career today:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">If I were a young person today, I would, without a doubt, aim to become the world&#8217;s best user of AI. I would be completely obsessed with how to use it, how to use different tools, open and closed models, everything. I would use all the different types of harnesses. I would learn how to manage its memory. How to create knowledge graphs, whether to use graphs. I would figure all of that out. I would become an expert at it. I would be constantly running new AI agents.</p>
</blockquote>



<p class="wp-block-paragraph">So, Huang believes young people should <strong>focus on mastering AI tools</strong>, dismissing predictions that more than half of young graduates will struggle to find jobs as ridiculous. </p>



<p class="wp-block-paragraph">And this advice is not limited to people working in tech. A plumber, for example, could use AI to help a customer design a new kitchen, while a furniture salesperson could offer a more detailed interior design proposal:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">By adopting technology, you can decide to do more with your own career, whatever your job may be.</p>
</blockquote>



<p class="wp-block-paragraph"><strong>That does not mean knowing how to use AI tools automatically replaces professional expertise</strong>. On the contrary, the more capable these tools become, the more important it is to evaluate their output, recognise mistakes and understand the problem they are supposed to solve. Someone who knows their field and understands how to apply AI within it has a tangible advantage over someone who relies solely on model outputs.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/11/Jensen-Huang-1024x614.jpg?x32039" alt="" class="wp-image-13278" srcset="https://shiftmag.dev/wp-content/uploads/2026/11/Jensen-Huang-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/11/Jensen-Huang-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/11/Jensen-Huang-767x460.jpg 767w, https://shiftmag.dev/wp-content/uploads/2026/11/Jensen-Huang.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>



<h2 class="wp-block-heading" id="ai-agents-from-different-companies-will-work-together">AI agents from different companies will work together</h2>



<p class="wp-block-paragraph">The conversation also turned to a future in which <strong>AI agents will play an increasingly important role in the workplace</strong>. Huang expects specialised agents to become more advanced and capable of handling increasingly complex tasks. One agent could research data, another analyse the findings, and a third prepare a report. A coordinating system would oversee their work and direct the individual steps.</p>



<p class="wp-block-paragraph">The NVIDIA CEO also <strong>expects agents from different systems and companies to start working together</strong>. This could connect, for example, industrial design tools with manufacturing platforms and business systems:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">An NVIDIA agent and a Mercedes agent, or an NVIDIA agent and a Siemens agent, could start working together once we connect them, whenever they determine that it is necessary.</p>
</blockquote>



<p class="wp-block-paragraph">But as agents gain more autonomy, <strong>the question of oversight</strong> becomes increasingly important. Huang highlighted security and access control as two of the key challenges in the development of AI agents. Systems should be given only the permissions necessary to complete a specific task, with clearly defined access rules and continuous monitoring of their activity.</p>



<p class="wp-block-paragraph">For companies, this means establishing data access policies, clear boundaries for agent autonomy and appropriate oversight mechanisms. An agent capable of carrying out tasks independently can improve efficiency, but mistakes or unintended actions can also have more serious consequences.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/11/Huang-2-1024x614.jpg?x32039" alt="" class="wp-image-13279" srcset="https://shiftmag.dev/wp-content/uploads/2026/11/Huang-2-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/11/Huang-2-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/11/Huang-2-767x460.jpg 767w, https://shiftmag.dev/wp-content/uploads/2026/11/Huang-2.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>



<h2 class="wp-block-heading" id="europe%e2%80%99s-ai-future"><strong>Europe’s AI future</strong></h2>



<p class="wp-block-paragraph">In Berlin, Huang also addressed Europe’s position in the global AI race. Europe does not need to develop every part of AI technology to become a relevant player, he argued. What matters more is identifying areas where it already has a competitive advantage, directing investment towards them and enabling new companies to emerge and apply AI in ways that were previously unimaginable:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Of course, you should aspire to do more. But, as with everything in life, you start by focusing on the areas where you are particularly good. You need to prioritise them because those are the areas where you are most likely to succeed first.</p>
</blockquote>



<p class="wp-block-paragraph">In Huang&#8217;s vision, Europe&#8217;s success over the next five years will be measured by <strong>how quickly existing companies integrate AI into their businesses</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Success means that every single company in Europe is an AI company.</p>
</blockquote>



<p class="wp-block-paragraph">That does not mean every company should develop its own AI model but <strong>integrating AI into products, services and business processes</strong>, as well as in the emergence of new companies that do not yet exist. <a href="https://www.konsulteer.com/article/nvidia-plans-investments-in-four-french-startups-as-france-s-sovereign-ai-push-widens" target="_blank" rel="noreferrer noopener">NVIDIA has already invested close to</a> $10 billion in Europe&#8217;s technology ecosystem, including startups, universities and research institutions:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">I believe in Europe&#8217;s AI startups. I believe they are Europe&#8217;s scouts. Not all of them have to succeed as independent companies. Some will find a sustainable business model, while others may be acquired by larger players. What matters is that they all have the opportunity to try.</p>
</blockquote>



<p class="wp-block-paragraph">Making that happen will also require a stronger venture capital ecosystem &#8211; Europe needs more investors willing to fund startups, as well as the capacity to support those investments.</p>



<h2 class="wp-block-heading" id="too-much-optimism"><strong>Too much optimism?</strong></h2>



<p class="wp-block-paragraph">Huang sees AI as an <strong>opportunity to expand human capabilities</strong>, develop new industries and create jobs that may not even exist today. At the same time, he acknowledges that this development comes with risks and that companies must take responsibility for the safety of their systems.</p>



<p class="wp-block-paragraph">His optimism, however, comes from the perspective of someone leading a company whose chips, systems and platforms power a large part of today&#8217;s AI infrastructure. NVIDIA benefits from a world in which more companies develop AI, invest in data centres and automate their operations. That does not necessarily make Huang&#8217;s predictions wrong, but it is worth considering the business interests behind the vision he presents.</p>



<p class="wp-block-paragraph">His vision describes a world in which AI can expand what individuals and companies are capable of achieving. Whether that opportunity translates into a genuine competitive advantage will also depend on who has the knowledge, resources and conditions needed to make the most of it.</p>



<p class="wp-block-paragraph"></p>
<p>The post <a href="https://shiftmag.dev/jensen-huangs-advice-to-young-developers-become-the-worlds-best-user-of-ai-13274/">Jensen Huang&#8217;s Advice to Young Developers: &#8220;Become the World&#8217;s Best User of AI&#8221;</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Only 7% of Organizations Using Kubernetes for AI Do So Daily</title>
		<link>https://shiftmag.dev/only-7-of-organizations-using-kubernetes-for-ai-do-so-daily-12086/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Tue, 15 Sep 2026 11:35:26 +0000</pubDate>
				<category><![CDATA[DevOps]]></category>
		<category><![CDATA[Event]]></category>
		<category><![CDATA[Infobip Shift 2026]]></category>
		<category><![CDATA[Kubernetes]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=12086</guid>

					<description><![CDATA[<p>Two-thirds of organizations run AI workloads on Kubernetes, but only 7% do it daily. Katie Gamanji says the ecosystem still needs better operational maturity and standards.</p>
<p>The post <a href="https://shiftmag.dev/only-7-of-organizations-using-kubernetes-for-ai-do-so-daily-12086/">Only 7% of Organizations Using Kubernetes for AI Do So Daily</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="1200" height="630" src="https://shiftmag.dev/wp-content/uploads/2026/09/katie_1.png?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/09/katie_1.png 1200w, https://shiftmag.dev/wp-content/uploads/2026/09/katie_1-300x158.png 300w, https://shiftmag.dev/wp-content/uploads/2026/09/katie_1-1024x538.png 1024w, https://shiftmag.dev/wp-content/uploads/2026/09/katie_1-768x403.png 768w" sizes="auto, (max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph">Yesterday at <a href="https://shiftmag.dev/tag/infobip-shift-2026/" target="_blank" rel="noreferrer noopener">Infobip Shift 2026</a>, I listened to Apple Principal Engineer and CNCF Technical Oversight Committee member Katie Gamanji talk about the <strong>state of cloud native and its slow turn towards AI</strong>.</p>



<p class="wp-block-paragraph">She said cloud native spent years becoming stable and unexciting, and AI teams now need that stability more than new tools. The big question is whether the people around that infrastructure can move as fast as the platform itself.</p>



<p class="wp-block-paragraph">For teams already using Kubernetes in production, the next change is very concrete, Katie said:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Dynamic Resource Allocation, or DRA, reached general availability in Kubernetes 1.34, adding a native way to request GPU and TPU capacity for AI workloads.</p>
</blockquote>



<h2 class="wp-block-heading" id="kubernetes-is-now-officially-boring">Kubernetes is now officially boring</h2>



<p class="wp-block-paragraph">Katie started with a number that captures how far Kubernetes has come: &#8220;98% of organizations in the CNCF’s annual report said they had adopted cloud native technologies.&#8221; After a decade of development, Kubernetes has become mainstream:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Kubernetes is now described as being boring, and I think this is a wonderful achievement.</p>
</blockquote>



<p class="wp-block-paragraph">The problem appears when AI workloads enter that mature infrastructure. According to the figures Katie presented, <strong>66% of organizations are already deploying AI workloads on Kubernetes</strong>. But only 7% are doing it daily, while 47% deploy them occasionally. Of the organizations using Kubernetes for AI, 23% have fully adopted the Kubernetes stack and 43% have partially adopted it.</p>



<p class="wp-block-paragraph">That 67 to 7 gap points to one likeliest cause. The talk did not name the cause directly, but the pattern Katie described points to operational immaturity as the likeliest one:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Teams fine-tune an existing model rather than build one, deploy it once to prove it works, and lack the automated retraining pipelines that would treat the model as a dynamic component.</p>
</blockquote>



<p class="wp-block-paragraph">The 7% deploying daily, she noted, &#8220;most likely implemented an automated retraining pipeline that treats the model as a dynamic component rather than a static element.&#8221; <strong>Running an AI workload is not as simple as putting another application into a container</strong>. Teams need model artifacts, rollback strategies, and training pipelines. Kubernetes is being positioned as the shared platform, though whether teams adopt it that way is another question.</p>



<h2 class="wp-block-heading" id="dra-and-ai-conformance-are-making-kubernetes-the-ai-platform">DRA and AI Conformance are making Kubernetes the AI platform</h2>



<p class="wp-block-paragraph">Katie compared this to DevOps: Kubernetes once brought developers and operations together, and now it could do the same for infrastructure teams and data scientists. But the analogy has limits &#8211; DevOps took years to work, and it is still not fully settled everywhere.</p>



<p class="wp-block-paragraph"><strong>Data scientists</strong>, for their part, <strong>rarely want to operate Kubernetes</strong>; they want a model served. Still, a platform that can run both traditional workloads and training or inference jobs reduces the need for separate AI infrastructure. </p>



<p class="wp-block-paragraph">Kubernetes is also adapting for this: DRA in 1.34 gives teams a more reliable way to allocate accelerators, which matters most for teams deploying daily.</p>



<p class="wp-block-paragraph">The <a href="https://www.cncf.io/announcements/2025/11/11/cncf-launches-certified-kubernetes-ai-conformance-program-to-standardize-ai-workloads-on-kubernetes/" target="_blank" rel="noreferrer noopener">AI Conformance Working Group</a> defines what a platform must support to run AI workloads reliably, with portability as the goal as Katie said: </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">If you have two conformant platforms, it&#8217;s going to be easy to lift and shift one product to the next platform.</p>
</blockquote>



<p class="wp-block-paragraph">Other initiatives target batch workloads, inference performance, and AI integration, and an Agent Sandbox effort is designing stateful, isolated runtimes for AI agents. The Serving Working Group completed its milestones and archived itself in February, continuing as a SIG on inference performance.</p>



<p class="wp-block-paragraph">The list is still changing quickly. Katie expects much of this landscape to look different within six months to a year.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="538" src="https://shiftmag.dev/wp-content/uploads/2026/09/katie_2-1024x538.png?x32039" alt="" class="wp-image-12131" srcset="https://shiftmag.dev/wp-content/uploads/2026/09/katie_2-1024x538.png 1024w, https://shiftmag.dev/wp-content/uploads/2026/09/katie_2-300x158.png 300w, https://shiftmag.dev/wp-content/uploads/2026/09/katie_2-768x403.png 768w, https://shiftmag.dev/wp-content/uploads/2026/09/katie_2.png 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption"> Photo: Petar Krajičić Vilović</figcaption></figure>



<h2 class="wp-block-heading" id="the-cloud-native-playbook-ai-can-borrow">The cloud native playbook AI can borrow</h2>



<p class="wp-block-paragraph">Katie divided the emerging open source AI ecosystem into three areas: <strong>training, inference, and agents</strong>. Training turns data into a model. Inference serves it. Agents connect it to the outside world.</p>



<p class="wp-block-paragraph">For platform leads, maturity matters. Training is led by the <a href="https://pytorch.org/foundation/" target="_blank" rel="noreferrer noopener">PyTorch Foundation</a>, inference has many options, and agents are still young under the <a href="https://aaif.io/" target="_blank" rel="noreferrer noopener">Agentic AI Foundation</a>.</p>



<p class="wp-block-paragraph"><strong>Cloud native experience can help AI tools mature faster</strong>. Security, observability, and identity were already solved in cloud native, and the CNCF project pipeline shows that kind of progress at scale.e.</p>



<p class="wp-block-paragraph">It also prunes: 28 archived projects. For anyone choosing a stack, the archive list is a practical filter, and Katie argued archival is a healthy sign, letting maintainers redirect energy toward projects that earn adoption.</p>



<h2 class="wp-block-heading" id="teams-that-engage-now-will-shape-the-patterns-everyone-else-follows">Teams that engage now will shape the patterns everyone else follows</h2>



<p class="wp-block-paragraph">For Katie, the next phase of cloud native is about whether the people building and maintaining the ecosystem can keep pace. She pointed out that contributing also means production feedback, feature requests, documentation, and white papers all shape projects.</p>



<p class="wp-block-paragraph">But in the end Katie&#8217;s talk left the hard questions unanswered, but the data shows why:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">98% of organizations trust Kubernetes with their infrastructure, two thirds have tried running AI on it, and almost nobody operates it continuously.</p>
</blockquote>



<p class="wp-block-paragraph">The specific bottlenecks, daily operational maturity, DRA adoption, and the conformance baseline, are still being defined, which means teams deploying AI on Kubernetes today are writing the patterns everyone else will copy:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">All of the working groups&#8217; meeting notes, invites, and repositories are public on GitHub. The teams that engage now, before the patterns harden, will not have to retrofit someone else&#8217;s choices in two years.</p>
</blockquote>



<p class="wp-block-paragraph"></p>
<p>The post <a href="https://shiftmag.dev/only-7-of-organizations-using-kubernetes-for-ai-do-so-daily-12086/">Only 7% of Organizations Using Kubernetes for AI Do So Daily</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>We Usually Blame the AI Model, But the Harness Is What Breaks Production</title>
		<link>https://shiftmag.dev/we-usually-blame-the-ai-model-but-the-harness-is-what-breaks-production-11032/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Thu, 06 Aug 2026 14:04:36 +0000</pubDate>
				<category><![CDATA[Artificial Intelligence]]></category>
		<category><![CDATA[Video]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[AWS]]></category>
		<category><![CDATA[developers]]></category>
		<category><![CDATA[Mike Chambers]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=11032</guid>

					<description><![CDATA[<p>Mike Chambers, Developer Advocate at AWS, says the real test of AI is not whether a demo looks smart, but whether the system around it can survive real-world use.</p>
<p>The post <a href="https://shiftmag.dev/we-usually-blame-the-ai-model-but-the-harness-is-what-breaks-production-11032/">We Usually Blame the AI Model, But the Harness Is What Breaks Production</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="1200" height="630" src="https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-10.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-10.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-10-300x158.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-10-1024x538.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-10-768x403.jpg 768w" sizes="auto, (max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph">A few years ago, building an AI app was hard. Today, anyone can put together an LLM, connect a few APIs, and build a strong demo in a day. But <strong>turning that demo into a production app is still much harder</strong>.</p>



<p class="wp-block-paragraph">At the AI Engineer World Fair, I talked with <strong>Mike Chambers</strong>, a Developer Advocate at AWS who focuses on generative AI, about what makes production AI systems different from conference demos. His answer kept coming back to one thing developers often overlook: everything around the model.</p>



<h2 class="wp-block-heading" id="its-not-ai-its-harness">It&#8217;s not AI, it&#8217;s harness</h2>



<p class="wp-block-paragraph">Developers spend a lot of time comparing models, tweaking prompts, and testing frameworks, but far less time on the infrastructure that keeps an AI app running. In his talk, Mike said many production failures don’t come from the model itself, they come from what he calls the <strong>harness</strong>.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">There&#8217;s two different types of agents: an agent that I&#8217;m using and an agent that I&#8217;m building, and we need to think about those very differently.</p>
</blockquote>



<p class="wp-block-paragraph"><br>For developers using coding assistants, the harness is mostly about standards, workflows, and how AI fits into the team’s development process. But for engineers building AI products, it means something completely different, as Mike put it:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">You have to think about how you&#8217;re actually going to deploy that out to infrastructure and deploy it at scale. And so doing that right and not just putting everything inside of one container, that&#8217;s the top tip here.</p>
</blockquote>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="538" src="https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-8-1024x538.jpg?x32039" alt="" class="wp-image-11054" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-8-1024x538.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-8-300x158.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-8-768x403.jpg 768w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-8.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>



<p class="wp-block-paragraph">That distinction matters because a lot of AI talk treats every agent as the same, but Mike says the engineering challenges are very different depending on whether you’re using an AI tool or building one for other people.</p>



<h2 class="wp-block-heading" id="observability-needs-to-be-built-in-early">Observability needs to be built in early</h2>



<p class="wp-block-paragraph">Today’s AI world is full of impressive demos. They can write code, answer questions, and automate tasks well enough to impress a crowd. But production systems need much more, so I asked Mike what teams usually get wrong when they move from demo to production:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">The number one thing is observability, followed closely by evaluations.</p>
</blockquote>



<p class="wp-block-paragraph">Mike said that used to be less important. Teams would build something cool, show it off, and stop there. Now, <strong>production AI needs much more discipline from the start</strong>.</p>



<p class="wp-block-paragraph">Modern AI systems, Mike says, work differently from traditional software. Their outputs can vary, their reasoning isn’t always easy to trace, and agents may take several steps to reach a result. That’s why observability needs to be built in early &#8211; so teams can understand what the system is doing before they try to ship it to production. </p>



<p class="wp-block-paragraph">Too often, AI projects are still judged mainly by model quality, when what really matters is whether the system is visible and controllable in practice.</p>



<h2 class="wp-block-heading" id="memory-doesnt-have-to-live-inside-the-agent">Memory doesn&#8217;t have to live inside the agent</h2>



<p class="wp-block-paragraph">Memory is one of the biggest topics in AI engineering, and it’s often treated as essential for useful agents. So I asked Mike what engineers misunderstand most about it.</p>



<p class="wp-block-paragraph">He said the biggest misconception is that an agent can’t work without memory. Short-term memory is just the current conversation, while long-term memory carries over into future interactions. The key question is where that processing happens:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">All of that memory extraction can happen outside of your agent somewhere else because you don&#8217;t need it today. You need it tomorrow when the user comes back to the agent again.</p>
</blockquote>



<figure class="wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio"><div class="wp-block-embed__wrapper">
<iframe loading="lazy" title="How AWS Thinks About Agent Harnesses" width="500" height="281" src="https://www.youtube.com/embed/ckZtEpZCkdA?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
</div></figure>



<p class="wp-block-paragraph">Rather than making every agent responsible for managing its own knowledge, <strong>developers can separate those concerns into different systems</strong>. That keeps agents simpler while still allowing applications to remember previous interactions.</p>



<h2 class="wp-block-heading" id="are-multi-agent-systems-the-future">Are multi-agent systems the future?</h2>



<p class="wp-block-paragraph">Another topic attracting plenty of attention are <strong>multi-agent architecture</strong>. Conference demos often showcase teams of specialized agents working together on different tasks. I was curious whether Mike sees that as the future of AI systems, so I asked him directly.</p>



<p class="wp-block-paragraph">He said that he doesn&#8217;t dismiss that approach, but he sees a different reason for using multiple agents in production. </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Multi-agent is super useful because you&#8217;re parallelizing the work. When developers build customer-facing AI systems, however, the motivation shifts. Often multi-agent is about separating out context. So it&#8217;s actually now becoming more of a context engineering conversation.</p>
</blockquote>



<p class="wp-block-paragraph">Instead of assigning agents to different jobs simply because it&#8217;s fashionable, teams should use them to <strong>isolate information and avoid overwhelming a single model</strong> with every piece of available context. The architecture becomes less about adding more intelligence and more about deciding which agent should see which information.</p>



<h2 class="wp-block-heading" id="engineers-need-to-build-a-strong-intuition-for-generative-ai-itself">Engineers need to build a strong intuition for generative AI itself</h2>



<p class="wp-block-paragraph">Mike’s most practical advice had little to do with frameworks or infrastructure. </p>



<p class="wp-block-paragraph">When I asked him what developers should focus on over the next few years, he didn’t point to another SDK or the latest model release. Instead, he argued that engineers need to build a strong intuition for generative AI itself &#8211; <strong>what large language models can do, and just as importantly, what they can’t</strong>. </p>



<p class="wp-block-paragraph">New models will keep arriving, but right now, he said, the biggest improvements are happening around the model, in the harness.</p>



<p class="wp-block-paragraph">His final observation neatly summarizes the <strong>engineering mindset</strong> he hopes developers adopt:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">You want to be using this technology. You don&#8217;t want this technology to be using you.</p>
</blockquote>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="538" src="https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-9-1-1024x538.jpg?x32039" alt="" class="wp-image-11058" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-9-1-1024x538.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-9-1-300x158.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-9-1-768x403.jpg 768w, https://shiftmag.dev/wp-content/uploads/2026/07/New-ShiftMag-panel-interview-9-1.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>
<p>The post <a href="https://shiftmag.dev/we-usually-blame-the-ai-model-but-the-harness-is-what-breaks-production-11032/">We Usually Blame the AI Model, But the Harness Is What Breaks Production</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Being a Senior Engineer Doesn’t Mean You Never Need Help</title>
		<link>https://shiftmag.dev/being-a-senior-engineer-doesnt-mean-you-never-need-help-10773/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Wed, 29 Jul 2026 13:46:11 +0000</pubDate>
				<category><![CDATA[Career]]></category>
		<category><![CDATA[Developer Productivity]]></category>
		<category><![CDATA[Infobip Shift 2026]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=10773</guid>

					<description><![CDATA[<p>The jump from mid-level to senior engineer is often framed as a technical one. Michelle Brenner says the real difference is judgment: understanding the business, making pragmatic trade-offs, and knowing when to ask for help.</p>
<p>The post <a href="https://shiftmag.dev/being-a-senior-engineer-doesnt-mean-you-never-need-help-10773/">Being a Senior Engineer Doesn’t Mean You Never Need Help</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="1200" height="630" src="https://shiftmag.dev/wp-content/uploads/2026/07/teresa.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/teresa.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/07/teresa-300x158.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/teresa-1024x538.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/teresa-768x403.jpg 768w" sizes="auto, (max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph">The jump from mid-level to senior engineer is usually described as a matter of stronger technical skills. </p>



<p class="wp-block-paragraph"><strong>Michelle Brenner</strong> (Senior Software Engineer) sees it differently. </p>



<p class="wp-block-paragraph">Ahead of her talk at <a href="https://shift.infobip.com/?%7CEU%7C-Shift_Zadar_2024-%5BTy:G_Brand_Search;Lng:EN;Bid:MAN_eCPC%5D" target="_blank" rel="noreferrer noopener">Infobip Shift</a>, (for which you get a <a href="https://www.entrio.hr/en/event/infobip-shift-2026-27033?pc=MAG15" target="_blank" rel="noreferrer noopener">special discount as a ShiftMag reader</a>) she argues that the real shift comes from <strong>understanding the business</strong>, making <strong>pragmatic decisions</strong>, and <strong>knowing when to ask for help</strong>.</p>



<h2 class="wp-block-heading" id="ai-makes-learning-at-work-easier">AI makes learning at work easier</h2>



<p class="wp-block-paragraph">You’re three weeks into a new job, the sprint is already underway, and a ticket comes in that doesn’t fully match the docs. Do you ask for help and maybe seem inexperienced, or do you guess and hope you got it right?</p>



<p class="wp-block-paragraph">That choice never really goes away. Senior engineers are not people who never need help, they <strong>know when to ask before uncertainty becomes a real problem</strong>.</p>



<p class="wp-block-paragraph">For self-taught engineers, that threshold can feel especially high, as Michelle describes it:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">As someone who was self-taught, I sometimes found that asking questions at work could be dangerous. I could reveal an ignorance that I was expected to know, from a computer science term to a tool name.</p>
</blockquote>



<p class="wp-block-paragraph">Michelle thinks <strong>AI can make it easier to ask basic questions</strong>. Since many developers already use AI tools connected to internal docs or code, they can get help without feeling judged.</p>



<p class="wp-block-paragraph">For her, that matters because most AI conversations focus on code generation. She cares more about what happens before that: helping engineers learn faster and feel less afraid to ask questions.</p>



<h2 class="wp-block-heading" id="seniors-also-need-business-context">Seniors also need business context</h2>



<p class="wp-block-paragraph">One of the biggest mistakes mid-level engineers make as they work toward becoming senior engineers, Michelle says, has little to do with syntax or frameworks. The real issue is <strong>not understanding how the business works</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Learn how the business makes money, and how your team affects that.</p>
</blockquote>



<p class="wp-block-paragraph">For Michelle, engineers who understand how leadership makes decisions and how a company creates value are better equipped to make technical calls that matter. So what can you do to fix that?</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">You should start paying more attention to business meetings and start seeing code as one part of a larger system, not the whole job.</p>
</blockquote>



<p class="wp-block-paragraph">In smaller teams, or in companies without a deep bench of specialists, that is often part of the job. Sometimes you have to take on <strong>the roles of product manager, developer, and SRE all at once</strong>, even when you are not the expert in every area.</p>



<h2 class="wp-block-heading" id="if-it-works-ship-it">&#8216;If it works, ship it&#8217;</h2>



<p class="wp-block-paragraph">That same pragmatism shapes how Michelle thinks about architecture and trade-offs.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">There is no single &#8220;correct&#8221; solution, just as there is no dream job or perfect candidate.</p>
</blockquote>



<p class="wp-block-paragraph">&#8220;If it works, ship it,&#8221; Michelle says. It is an argument against <strong>getting stuck in the search for a perfect answer that does not exist</strong>. Senior engineers spend less time chasing theoretical purity and more time balancing constraints, risks, and deadlines.</p>



<p class="wp-block-paragraph">Michelle also gave us a sneak peek of her upcoming talk at <a href="https://shift.infobip.com/?%7CEU%7C-Shift_Zadar_2024-%5BTy:G_Brand_Search;Lng:EN;Bid:MAN_eCPC%5D" target="_blank" rel="noreferrer noopener">Infobip Shift</a>, based on the senior engineering career guide she has spent the last few years writing. She says the jump from mid to senior is much bigger than the jump from junior to mid, and that it can be <strong>especially hard without a mentor</strong> to show how the role actually works.</p>



<p class="wp-block-paragraph">Her talk will focus on the <strong>practical skills engineers often do not get enough time to learn on the job</strong>, from build-versus-buy decisions to interview preparation. She also spoke with engineers from organisations around the world to keep the advice broad. If you are coming to Shift, this is one talk worth catching.</p>



<p class="wp-block-paragraph"><strong>Get your ticket for Shift conference with a&nbsp;<a href="https://www.entrio.hr/en/event/infobip-shift-2026-27033?pc=MAG15" target="_blank" rel="noreferrer noopener">special discount for ShiftMag readers</a>!</strong></p>
<p>The post <a href="https://shiftmag.dev/being-a-senior-engineer-doesnt-mean-you-never-need-help-10773/">Being a Senior Engineer Doesn’t Mean You Never Need Help</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>PewDiePie’s AI Experiment Is What Software Development Looks Like Now</title>
		<link>https://shiftmag.dev/pewdiepies-ai-experiment-is-what-software-development-looks-like-now-10697/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Tue, 28 Jul 2026 12:48:08 +0000</pubDate>
				<category><![CDATA[Tools]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[local AI]]></category>
		<category><![CDATA[Odysseus AI]]></category>
		<category><![CDATA[self-hosted]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=10697</guid>

					<description><![CDATA[<p>I let the dust settle, and this is what I think: regular apps usually need teams of dozens to ideate, build, and test before they ship. Odysseus, PewDiePie’s new AI platform, doesn’t - at least, not in the usual way.</p>
<p>The post <a href="https://shiftmag.dev/pewdiepies-ai-experiment-is-what-software-development-looks-like-now-10697/">PewDiePie’s AI Experiment Is What Software Development Looks Like Now</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="1200" height="630" src="https://shiftmag.dev/wp-content/uploads/2026/07/JPA.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/JPA.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/07/JPA-300x158.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/JPA-1024x538.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/JPA-768x403.jpg 768w" sizes="auto, (max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph">Odysseus is interesting because it shows <strong>how a project with serious ambitions can now be built outside a traditional software company</strong>. </p>



<p class="wp-block-paragraph">After speaking with <strong>Ivan</strong> <strong>Mihić</strong>, a software engineer who spent time testing PewDiePie&#8217;s AI project, I ended up thinking that the software itself might not be the most interesting part of this story. </p>



<p class="wp-block-paragraph"><a href="https://odysseusai.dev/" target="_blank" rel="noreferrer noopener">Odysseus</a> may come from a YouTube creator with a massive audience, but the more important question is what it says about software development itself. </p>



<h2 class="wp-block-heading" id="another-ai-app-or-something-more">Another AI app &#8211; or something more?</h2>



<p class="wp-block-paragraph">When <a href="https://youtu.be/rAzT5lcezPs" target="_blank" rel="noreferrer noopener">Felix &#8220;PewDiePie&#8221; Kjellberg unveiled Odysseus</a>, the internet reacted in the way it usually does when a major creator launches a software project. Some treated it as a celebrity side project, others focused on the privacy claims and some compared it with other AI tools such as GPT or Claude. </p>



<p class="wp-block-paragraph">For those of you who might be late to the party (same as we are with this text), Odysseus is an <strong>open source, self hosted AI workspace designed to run locally</strong>. Instead of depending only on cloud services, it combines local models with web search, document analysis, long term memory, and AI agents inside a single browser interface. If you want, you can also add cloud models to it as well.</p>



<p class="wp-block-paragraph">Under the hood, the application integrates components such as&nbsp;ChromaDB&nbsp;for memory,&nbsp;SearXNG&nbsp;for search, and local model serving, while also letting users decide whether they want everything on their own hardware or connected to external APIs.&nbsp;</p>



<p class="wp-block-paragraph">What stood out to Ivan was not just what Odysseus does, but what its existence says about <strong>how software gets built nowadays</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">The biggest thing isn&#8217;t that Odysseus is a good tool,&nbsp;It&#8217;s&nbsp;that a project of this size probably wouldn&#8217;t exist without AI.</p>
</blockquote>



<p class="wp-block-paragraph">A few years ago, building a system that combined models, search, memory, document pipelines, and agents would usually&nbsp;have required&nbsp;a much larger engineering team. Today, a determined individual or a small group can at least prototype that kind of product and keep iterating on it. </p>



<p class="wp-block-paragraph">To me as an outsider, this looks like the <strong>phase in which we built AI tools mainly for internal use is slowly coming to an end</strong>. More and more of these tools are now being prepared for public release, even if they are still far from fully polished.</p>



<p class="wp-block-paragraph">Ivan also&nbsp;admints&nbsp;that he was&nbsp;was&nbsp;surprised. Not because the tool is&nbsp;perfect,&nbsp;because it as you would expect isn&#8217;t&nbsp;but because something that started from a hobbyist, vibe-coding context grew into a project of this scale. For him, Odysseus is less a sign that AI has replaced engineers and more a sign that AI has changed the starting line. And I couldn&#8217;t agree more on this one.</p>



<h2 class="wp-block-heading" id="privacy-alone-won%e2%80%99t-protect-you%c2%a0">Privacy alone won’t protect you </h2>



<p class="wp-block-paragraph">Much of Odysseus’ appeal comes from privacy. Running AI models locally means <strong>sensitive data does not have to leave your machine</strong>, which is a real advantage. But Ivan pointed out that privacy and security are not the same thing. </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Odysseus is local-first, but that doesn&#8217;t automatically make it secure.&nbsp;</p>
</blockquote>



<p class="wp-block-paragraph">The platform can access files, execute shell commands, perform web research, and interact with external services, therefore a bad configuration can quickly create security risks. That matters even more if users expose their instance to the public internet.&nbsp;</p>



<p class="wp-block-paragraph">Ivan’s point is simple: <strong>a tool can be private in design and still be unsafe in practice</strong> if people do not understand what they are connecting, exposing, or automating. </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">When a tool presents itself through the lens of privacy, security becomes extremely important. Privacy without security&nbsp;doesn&#8217;t&nbsp;mean much.</p>
</blockquote>



<h2 class="wp-block-heading" id="odysseus-is-not-mature-yet">Odysseus is not mature&#8230; yet</h2>



<p class="wp-block-paragraph">Architecturally, Ivan sees Odysseus as a <strong>large monolithic application coordinating several supporting services</strong> rather than a fully mature enterprise platform. </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">It&#8217;s not the cleanest or most mature architecture I&#8217;ve seen. You can clearly&nbsp;tell&nbsp;it evolved through experimentation.</p>
</blockquote>



<p class="wp-block-paragraph">That <strong>experimental nature is visible throughout the product</strong>. Odysseus tries to combine chat, agents, research, memory, document analysis, and productivity tools into one personal AI workspace, with varying degrees of success.</p>



<p class="wp-block-paragraph">A few features stood out to Ivan in particular.&nbsp;&#8220;Compare&#8221; lets users run the same prompt across multiple models at once. &#8220;Deep Research&#8221; automatically performs multi step web research and returns cited sources.&nbsp;&#8220;Cookbook&#8221; helps users choose, download, and serve models based on the hardware they actually have.&nbsp;</p>



<p class="wp-block-paragraph">Still, for someone who is just entering the world of local AI, Ivan thinks more established tools currently make more sense, just because tools like Open WebUI has had more time to mature.</p>



<h2 class="wp-block-heading" id="this-is-not-exactly-a-beginner-friendly-tool">This is not exactly a beginner-friendly tool</h2>



<p class="wp-block-paragraph">Talking about Odysseus once it&#8217;s installed is fine, but <strong>getting there as a regular AI user might prove to be a challenging task</strong>. From my experience, users need to be wary of the fact that, in order to fully use Odysseus, many setup steps are needed. They also need to know that they might not understand a single thing about what&#8217;s happening, which is also a risk in itself.</p>



<p class="wp-block-paragraph">Odysseus is not an app you can install and run in a minute or two, while other tools do just that. The user experience for a beginner might be <strong>overwhelming</strong> and, at times, just simply not worth it if you&#8217;re in it just to experiment and see what the tool can offer.</p>



<p class="wp-block-paragraph">Ivan Šimić, the third Ivan in this story, a senior writer at ShiftMag who kept experimenting with the tool to figure it out, added:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Trying to hop on the bandwagon of local LLM&#8217;s after reading about Odysseus might not be the greatest idea, since it requires a capable machine, a basic understanding of key aspects of making things happen on a PC or Mac, patience and time. For many interested in trying it, this might prove to be too much &#8220;tinkering&#8221; for not a lot of results. </p>
</blockquote>



<h2 class="wp-block-heading" id="big-ideas-in-software-are-no-longer-a-big-team-privilege">Big ideas in software are no longer a big-team privilege</h2>



<p class="wp-block-paragraph">It would be easy to frame Odysseus as another celebrity tech experiment but that would miss the broader point.&nbsp;</p>



<p class="wp-block-paragraph">What matters here is that <strong>AI assisted development has lowered the cost of building ambitious software</strong>. A project that once would have required a sizable engineering team is now something a small group, or even one person, can assemble and refine much faster than before. </p>



<p class="wp-block-paragraph">And yes, while a fair portion of the pessimistic internet is claiming this means software engineers are disappearing, I highly doubt it. The job is changing, and the gap between an idea and a working product is smaller than it used to be.  </p>



<p class="wp-block-paragraph">As Ivan nicely put it, it’s like Jack Black’s quest for the magical guitar pick in The Pick of Destiny. In the film, that pick gives its owner superhuman guitar skills. In reality, <strong>even the best tool won’t make him or anybody else the world’s best developer</strong>, but it can certainly make everyday job a lot easier.</p>



<p class="wp-block-paragraph">Therefore, the most interesting thing PewDiePie built may not be another AI assistant, but proof that ambitious software is no longer reserved for traditional software companies and big teams. Big Tech, are you scared yet?</p>
<p>The post <a href="https://shiftmag.dev/pewdiepies-ai-experiment-is-what-software-development-looks-like-now-10697/">PewDiePie’s AI Experiment Is What Software Development Looks Like Now</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Instagram&#8217;s Co-Founder Says He Doesn&#8217;t Review Every Line of Code Anymore</title>
		<link>https://shiftmag.dev/instagrams-co-founder-says-he-doesnt-review-every-line-of-code-anymore-11063/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Mon, 27 Jul 2026 12:48:51 +0000</pubDate>
				<category><![CDATA[Artificial Intelligence]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[AI Engineer]]></category>
		<category><![CDATA[developers]]></category>
		<category><![CDATA[Engineers]]></category>
		<category><![CDATA[Mike Krieger]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=11063</guid>

					<description><![CDATA[<p>At the AI Engineer conference in San Francisco, I heard Instagram co-founder Mike Krieger say developers no longer need to review every line of code by hand. The real job now is judging AI’s decisions and whether the product solves the right problem.</p>
<p>The post <a href="https://shiftmag.dev/instagrams-co-founder-says-he-doesnt-review-every-line-of-code-anymore-11063/">Instagram&#8217;s Co-Founder Says He Doesn&#8217;t Review Every Line of Code Anymore</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="1200" height="720" src="https://shiftmag.dev/wp-content/uploads/2026/07/Mike-1.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/Mike-1.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-1-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-1-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-1-768x461.jpg 768w" sizes="auto, (max-width: 1200px) 100vw, 1200px" /></figure>


<p class="wp-block-paragraph"><strong>Mike Krieger</strong>, Instagram co-founder and Member of Technical Staff at Anthropic, said hand-reviewing every line of code is no longer the best use of a developer’s time. </p>



<p class="wp-block-paragraph">The job, he argued, is moving toward <strong>understanding what AI is trying to do</strong>, judging whether its decisions make sense, and checking whether the end result actually solves the user’s problem. </p>



<p class="wp-block-paragraph">For startups and small teams, that shift could be a major advantage: AI makes it cheaper to experiment, pivot, and learn faster from real feedback.</p>



<h2 class="wp-block-heading" id="ai-will-not-replace-developers">&#8216;AI will not replace developers&#8217;<br></h2>



<p class="wp-block-paragraph">If a founder (or a developer) can test three or four product versions or approaches in the time it used to take to build one, <strong>the biggest gain is speed of learning</strong>. </p>



<p class="wp-block-paragraph">In Mike&#8217;s view, that doesn’t mean AI replaces creativity, ideation, or product instinct, it just helps strong ideas get validated, or rejected, much faster.</p>



<p class="wp-block-paragraph">The real challenges developers and founders face were never technical to begin with, he argues. Spotting the actual problem, <strong>understanding what users need</strong>, and finding a product the market will genuinely adopt is far harder than writing the code itself. </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">AI will not replace developers or threaten the startup ecosystem. It will let small teams build, test, and iterate faster with fewer resources.<br></p>
</blockquote>



<p class="wp-block-paragraph">That shift will not be easy for everyone, and <strong>plenty of teams will struggle</strong>, quit, or pivot along the way. But blaming AI for a rough patch in the industry is not fair. Because as Mike says, what was hard before AI is still hard now.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/07/Mike-2-1024x614.jpg?x32039" alt="" class="wp-image-11067" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/Mike-2-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-2-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-2-768x461.jpg 768w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-2.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>



<h2 class="wp-block-heading" id="code-review-matters-less-now">Code review matters less now<br></h2>



<p class="wp-block-paragraph">One of the biggest changes Mike sees is not how much code AI can write, but how people assign work to it.</p>



<p class="wp-block-paragraph">Developers are moving away from planning every step in detail and doing it all by hand. Now they more often<em> </em><strong>set the goal, give the AI context, and then review the results</strong>, decisions, and trade-offs.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">At Anthropic AI is increasingly used less as a simple tool for individual tasks and more like a team member that holds context, tracks specific problems, and proactively suggests solutions.</p>
</blockquote>



<p class="wp-block-paragraph">That is also changing how code gets reviewed. Traditional code review, where someone goes line by line and checks the implementation, matters less than understanding the intent behind a change.</p>



<p class="wp-block-paragraph">Instead of asking &#8220;is every line of this code correct,&#8221; the more useful question becomes &#8220;<strong>is this the right way to solve the problem, and what are the consequences of this decision</strong>.&#8221;</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">I don&#8217;t review every line of code in a pull request. Instead, I talk to Claude about the code and ask the questions I would normally ask the person who wrote it.</p>
</blockquote>



<p class="wp-block-paragraph">That does not mean people disappear from the process. Human judgment matters more, especially for <strong>architectural decisions and complex changes</strong>. AI can speed up implementation, but someone still needs to understand the product, the users, and the long-term consequences of a decision.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/07/Mike-3-1024x614.jpg?x32039" alt="" class="wp-image-11068" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/Mike-3-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-3-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-3-768x461.jpg 768w, https://shiftmag.dev/wp-content/uploads/2026/07/Mike-3.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>



<h2 class="wp-block-heading" id="it-is-easy-to-lose-perspective-in-the-ai-race">It is easy to lose perspective in the AI race<br></h2>



<p class="wp-block-paragraph">Beyond the changing workflow, Mike also spoke about the pressure in the AI industry. With new models, products, and regulatory shifts coming almost daily, it is easy to feel like you always have to stay one step behind.</p>



<p class="wp-block-paragraph">He said that by midweek, Anthropic is already reviewing the ”state of AI news,” and the list of updates is often long. His point was that <strong>you have to keep perspective</strong>, you can’t track or react to every single change.</p>



<p class="wp-block-paragraph">That pressure, he added, is also what makes <strong>burnout</strong> such a risk in an industry moving this fast. People often tie their sense of success to daily results, competitors’ launches, or product releases, even though real wins and failures are usually measured over much longer stretches of time.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">No job is important enough that you can&#8217;t be offline for a few days.</p>
</blockquote>



<p class="wp-block-paragraph">He closed by saying <strong>teams need to get used to talking about frustrations and failures, not just celebrating wins</strong>. Being open about problems, he argued, helps teams solve them faster and keep a healthy working culture, even as the industry moves at an extraordinary pace.</p>
<p>The post <a href="https://shiftmag.dev/instagrams-co-founder-says-he-doesnt-review-every-line-of-code-anymore-11063/">Instagram&#8217;s Co-Founder Says He Doesn&#8217;t Review Every Line of Code Anymore</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>It May Take Longer to Review a PR Than It Takes to Write It</title>
		<link>https://shiftmag.dev/it-may-take-longer-to-review-a-pr-than-it-takes-to-write-it-9927/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Fri, 17 Jul 2026 14:06:11 +0000</pubDate>
				<category><![CDATA[Productivity]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[AI code review]]></category>
		<category><![CDATA[developer tools]]></category>
		<category><![CDATA[skill erosion]]></category>
		<category><![CDATA[software engineering]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=9927</guid>

					<description><![CDATA[<p>That mismatch is becoming the real cost of AI-assisted development: code ships faster, but the work of understanding it, checking it, and trusting it hasn’t sped up at all.</p>
<p>The post <a href="https://shiftmag.dev/it-may-take-longer-to-review-a-pr-than-it-takes-to-write-it-9927/">It May Take Longer to Review a PR Than It Takes to Write It</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="2100" height="1413" src="https://shiftmag.dev/wp-content/uploads/2026/07/INF09895-scaled.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/INF09895-scaled.jpg 2100w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09895-300x202.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09895-1024x689.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09895-768x517.jpg 768w" sizes="auto, (max-width: 2100px) 100vw, 2100px" /></figure>


<p class="wp-block-paragraph"><strong>Olena Babenko</strong>, Staff Software engineer at Aiven, cut straight to the uncomfortable truth in a recent interview I had with her in London.</p>



<p class="wp-block-paragraph">The oversight model collapses the moment the person approving the output can <strong>no longer reliably tell whether it’s good</strong>, bad, or just convincingly wrong.</p>



<p class="wp-block-paragraph">She pointed to a growing mismatch between how fast code is being generated and how slowly it is actually being checked.</p>



<h2 class="wp-block-heading" id="ai-is-speeding-up-code-creation-but-review-can%e2%80%99t-keep-up">AI is speeding up code creation, but review can’t keep up</h2>



<p class="wp-block-paragraph">Olena says she keeps running into the same cycle, that looks like this &#8220;<strong>a steady stream of announcements about how the newest model will change software engineering</strong>, followed almost weekly by another company unveiling its most powerful release yet&#8221;.</p>



<p class="wp-block-paragraph">Then someone actually turns the model on and gives it a simple, but domain-specific task, and it hallucinates, a lot</p>



<p class="wp-block-paragraph">She sees this happen often enough that the pattern itself has become the story. The industry isn&#8217;t there yet, she says, at least not in the systems she works with.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="689" src="https://shiftmag.dev/wp-content/uploads/2026/07/INF09862-1-1024x689.jpg?x32039" alt="" class="wp-image-10757" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/INF09862-1-1024x689.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09862-1-300x202.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09862-1-768x517.jpg 768w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Marin Pavelić</figcaption></figure>



<p class="wp-block-paragraph">The real problem, in her view, shows up one step later in the pipeline: engineers are spending less time writing code and more time reviewing it, and the incentive structures around that shift have not caught up.</p>



<p class="wp-block-paragraph">Therefore, I asked her what that imbalance looks like:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">We have an ability to generate code very, very fast, thousands of lines of code, and without thinking how people perceive that code, how readable it is, how good it is, how simple it is to understand, but code reviewing, scaling, and automation is not on that same level yet.</p>
</blockquote>



<p class="wp-block-paragraph">As a result, you can create a PR, and a person who actually reviews it took <strong>much more time to review it than you spent writing it</strong>. This means that the problem, at least according to Olena, is structural, not just technical:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">The current system and a lot of managerial tools are made on KPIs, they are made on individual contributions of code writing, so if you look at the GitHub contributions or Jira tickets, they are praising writing code, finishing tasks, creating PRs, while the people who do the PR reviews, their jobs are not valued that much.</p>
</blockquote>



<p class="wp-block-paragraph">Add unusually verbose AI-generated code into that mix, and reviewing takes even longer than the writing did.</p>



<h2 class="wp-block-heading" id="the-review-burden-has-always-been-there-but-it%e2%80%99s-getting-worse">The review burden has always been there, but it’s getting worse</h2>



<p class="wp-block-paragraph">Olena is careful to point out this isn&#8217;t new. She says that a reviewer burden existed even before all this and <strong>AI is just amplifying this problem</strong>.</p>



<p class="wp-block-paragraph">She describes a colleague&#8217;s experience at a previous company: engineers spending two weeks on an issue, then dropping a thousand-line PR on a reviewer who was, realistically, expected to just hit merge, not actually engage with the work. </p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="689" src="https://shiftmag.dev/wp-content/uploads/2026/07/INF09872-1024x689.jpg?x32039" alt="" class="wp-image-10758" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/INF09872-1024x689.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09872-300x202.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/INF09872-768x517.jpg 768w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Marin Pavelić</figcaption></figure>



<p class="wp-block-paragraph">The fix, in her view, is not more automation but <strong>earlier communication, smaller PRs, and separating refactoring from actual logic changes</strong>, paired with management that treats reviewing as real work worth rewarding:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">This is a question of management.</p>
</blockquote>



<h2 class="wp-block-heading" id="ai-can-make-tests-look-good-without-actually-making-the-code-reliable">AI can make tests look good without actually making the code reliable</h2>



<p class="wp-block-paragraph">Olena also pointed to one of the first things AI tends to break in a codebase: <strong>unit tests</strong>. Writing them is boring, so people delegate it a lot, but AI can already game the system, making tests pass without making the code reliable.</p>



<p class="wp-block-paragraph">By her account, models will patch internal implementations or hardcode expected outputs rather than genuinely verify behavior, leaving teams with high coverage numbers that protect against nothing &#8211; though the damage doesn’t show up right away.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">They are not falling apart in one day. This is a process that&#8217;s eroding your system with a small step day by day until you realize that your new system is really unreliable, really something you&#8217;re not supposed to do in production.</p>
</blockquote>



<p class="wp-block-paragraph">By the time anyone notices, <strong>the damage started much earlier</strong>.</p>



<h2 class="wp-block-heading" id="we-are-delegating-too-much-thinking-to-ai">&#8216;We are delegating too much thinking to AI&#8217;</h2>



<p class="wp-block-paragraph">Olena&#8217;s biggest long-term concern is <strong>skill erosion</strong>:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">We’re delegating too much thinking to AI. But AI doesn’t become an expert just because we use it: it has no memory, it can’t really learn, and it can’t build experience. Humans can do that, at least for now. The problem is that we’re not really learning from the experience either, so our skills and critical thinking keep getting weaker.</p>
</blockquote>



<p class="wp-block-paragraph">That undercuts the standard reassurance that a human stays in the loop. &#8220;You might end up in a situation where you have a human in the loop, but those people do not have correct skills anymore&#8221;, Olena said.</p>



<p class="wp-block-paragraph"> A safety net only works if the person holding it still knows how to catch something.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="768" src="https://shiftmag.dev/wp-content/uploads/2026/07/Olena-Babenko-i-Pelin-2-1024x768.jpg?x32039" alt="" class="wp-image-10759" srcset="https://shiftmag.dev/wp-content/uploads/2026/07/Olena-Babenko-i-Pelin-2-1024x768.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/07/Olena-Babenko-i-Pelin-2-300x225.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/07/Olena-Babenko-i-Pelin-2-768x576.jpg 768w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Ivan Pelivanović</figcaption></figure>
<p>The post <a href="https://shiftmag.dev/it-may-take-longer-to-review-a-pr-than-it-takes-to-write-it-9927/">It May Take Longer to Review a PR Than It Takes to Write It</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Why Good Architecture Still Fails in Production</title>
		<link>https://shiftmag.dev/why-good-architecture-still-fails-in-production-9894/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Tue, 30 Jun 2026 12:52:02 +0000</pubDate>
				<category><![CDATA[Backend]]></category>
		<category><![CDATA[Engineering Management]]></category>
		<category><![CDATA[development]]></category>
		<category><![CDATA[Devoxx]]></category>
		<category><![CDATA[software architecture]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=9894</guid>

					<description><![CDATA[<p>At Devoxx UK in London, I caught up with Eoin Woods - co-author of three software architecture books and former CTO of Endava, to find out why beautifully designed systems still break in production.</p>
<p>The post <a href="https://shiftmag.dev/why-good-architecture-still-fails-in-production-9894/">Why Good Architecture Still Fails in Production</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="2100" height="1400" src="https://shiftmag.dev/wp-content/uploads/2026/05/Eoin-devoxx-2-1-scaled.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/05/Eoin-devoxx-2-1-scaled.jpg 2100w, https://shiftmag.dev/wp-content/uploads/2026/05/Eoin-devoxx-2-1-300x200.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/05/Eoin-devoxx-2-1-1024x683.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/05/Eoin-devoxx-2-1-768x512.jpg 768w" sizes="auto, (max-width: 2100px) 100vw, 2100px" /></figure>


<p class="wp-block-paragraph">Eoin, who now works independently across software architecture, green software, and engineering, <strong>has spent decades around large systems</strong>. </p>



<p class="wp-block-paragraph">The failures he keeps seeing are rarely exotic. They begin with something mundane, an environmental detail nobody verified, a technology nobody fully understood, or a second order effect nobody expected until the system was already live.</p>



<p class="wp-block-paragraph">That is why, for him, architecture is less about drawing clean systems on paper and more about <strong>surfacing the assumptions production will eventually expose</strong>.</p>



<h2 class="wp-block-heading" id="production-is-where-the-assumptions-become-visible">Production is where the assumptions become visible</h2>



<p class="wp-block-paragraph">When I asked Eoin where failures come from, he resisted the temptation to separate elegant design from messy execution. He argued, the system that matters is the one that ships, and the one that ships is full of assumptions.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Ultimately the problem is always the design or actually, not even the design but what ends up in production&#8230;</p>
</blockquote>



<p class="wp-block-paragraph">This means that diagram may look reasonable, the architecture review may pass and the slide deck may even sound convincing. Production is where the assumptions become visible. <strong>Experience changes what people notice</strong>, Eoin put this bluntly:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">The older and more cynical you get, the more questions you ask.</p>
</blockquote>



<p class="wp-block-paragraph">Seniority, at least in architecture, often means <strong>noticing the missing question</strong> before the missing answer turns into an incident. The same logic applies to coupled systems. A component can behave well on its own and still create trouble once the rest of the system starts reacting to it.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="683" src="https://shiftmag.dev/wp-content/uploads/2026/06/Eoin-devoxx-1-1024x683.jpg?x32039" alt="" class="wp-image-10205" srcset="https://shiftmag.dev/wp-content/uploads/2026/06/Eoin-devoxx-1-1024x683.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/06/Eoin-devoxx-1-300x200.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/06/Eoin-devoxx-1-768x512.jpg 768w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Marin Pavelić</figcaption></figure>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">If one component slows down dramatically, the others (if tightly bound to it) probably slow down dramatically as well. It’s only when it actually happens that everyone realises everything else slows down in lockstep, because they hadn’t realised how coupled together they were.</p>
</blockquote>



<p class="wp-block-paragraph">That is the kind of failure teams usually understand only after the fact. The system behaves as designed, then fails as a whole because the interactions were never really understood as interactions.</p>



<h2 class="wp-block-heading" id="architecture-is-never-just-about-structure">Architecture is never just about structure</h2>



<p class="wp-block-paragraph">That distinction matters most where recovery is expensive. Financial infrastructure, healthcare, industrial control, and similar environments do not get the luxury of treating failure as a learning exercise.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">We’re much less enthusiastic about moving fast, breaking things and fixing forwards actually because if we do break something in a mission critical system it has a really serious side effect.</p>
</blockquote>



<p class="wp-block-paragraph">In those environments, <strong>speed only helps when recovery is cheap</strong>. If rollback is hard, repair is slow, or the impact is irreversible, then fast delivery stops being a virtue and starts becoming a liability.</p>



<p class="wp-block-paragraph">That is why architecture is never just about structure, but about consequences. A design that looks efficient in a presentation can become expensive in the real world the moment its assumptions meet a system that cannot absorb mistakes.</p>



<h2 class="wp-block-heading" id="writing-the-scenario-forces-the-hidden-question-into-the-open">Writing the scenario forces the hidden question into the open</h2>



<p class="wp-block-paragraph">Eoin made the same argument in his Devoxx talk by describing a scene almost every engineer will recognise. </p>



<p class="wp-block-paragraph">A stakeholder asks for something scalable, cost effective, secure, and easy to use. The architect comes back with containerised microservices, lower storage costs, forced password changes, and a task oriented interface. The stakeholder says it sounds good, while still not really understanding what was asked or what was decided.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="683" src="https://shiftmag.dev/wp-content/uploads/2026/06/55258320547_374de78c6f_o-1024x683.jpg?x32039" alt="" class="wp-image-10203" srcset="https://shiftmag.dev/wp-content/uploads/2026/06/55258320547_374de78c6f_o-1024x683.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/06/55258320547_374de78c6f_o-300x200.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/06/55258320547_374de78c6f_o-768x512.jpg 768w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Devoxx / Flickr</figcaption></figure>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">That gap between what was asked and what was understood is exactly what architectural scenarios are designed to close.</p>
</blockquote>



<p class="wp-block-paragraph">A scenario takes a vague wish and turns it into something concrete. Not &#8220;the system should be scalable&#8221; but what happens when 5,000 users connect at the same time and the primary database fails at 7pm during peak load. Not &#8220;the app should be secure&#8221; but what happens when a decryption key needs to be recovered after an incident. The value of that exercise is not subtle.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">There’s nothing like writing a scenario to reveal what you don’t know about your own system.</p>
</blockquote>



<p class="wp-block-paragraph">That is why scenarios matter, Eoin believes. They do not make uncertainty disappear, but they do force it into view.</p>



<p class="wp-block-paragraph">He described six practical uses for scenarios. Teams use them to:</p>



<ul class="wp-block-list">
<li>decide what to build</li>



<li>compare design options</li>



<li>drive research</li>



<li>assess design choices more broadly</li>



<li>explain existing system behaviour to people who need to understand it without getting lost in technical detail</li>



<li>surface the questions that should have been asked much earlier</li>
</ul>



<p class="wp-block-paragraph">The point is not documentation for its own sake though, but to <strong>make the system legible</strong> before production does the explanation instead.</p>



<p class="wp-block-paragraph">If offline mode has only been demonstrated and never truly tested against the live system, the scenario is where that becomes obvious. If a recovery flow exists mainly in a vendor demo, the scenario turns the demo into a question rather than a conclusion.</p>



<h2 class="wp-block-heading" id="the-simplest-useful-habit-is-still-discipline">The simplest useful habit is still discipline</h2>



<p class="wp-block-paragraph">Eoin’s most practical advice was also the least glamorous: </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Make architectural decisions intentionally, and write them down.</p>
</blockquote>



<p class="wp-block-paragraph">This means <strong>recording the assumptions built into it, the trade offs it accepts, and the implications that follow</strong>.</p>



<p class="wp-block-paragraph">ADRs, or <strong>architecture decision records</strong>, are not a new idea. Eoin said he was already writing them as a graduate in the 1990s, and even then they were not common practice. The fact that the advice is old does not make it less relevant. If anything, the opposite is true.</p>



<p class="wp-block-paragraph">AI tooling is generating code faster than teams can inspect the assumptions that end up inside it. Requirements are still vague. Stakeholders still do not always understand the technical response and decisions are still too often left implicit.</p>



<p class="wp-block-paragraph">The result is more <strong>software with more hidden assumptions inside it</strong>.</p>



<p class="wp-block-paragraph">That is where architecture really begins, in the part of the work that makes those assumptions visible before production has to do it for you.</p>



<p class="wp-block-paragraph"></p>



<p class="wp-block-paragraph"></p>



<p class="wp-block-paragraph"></p>
<p>The post <a href="https://shiftmag.dev/why-good-architecture-still-fails-in-production-9894/">Why Good Architecture Still Fails in Production</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Trisha Gee: AI Won&#8217;t Fix Your Broken Pipeline &#8211; It Will Break It Faster</title>
		<link>https://shiftmag.dev/trisha-gee-ai-wont-fix-your-broken-pipeline-it-will-break-it-faster-9785/</link>
		
		<dc:creator><![CDATA[Ivan Pelivanovic]]></dc:creator>
		<pubDate>Wed, 27 May 2026 13:36:39 +0000</pubDate>
				<category><![CDATA[Artificial Intelligence]]></category>
		<category><![CDATA[Video]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[Developer Productivity]]></category>
		<category><![CDATA[development]]></category>
		<guid isPermaLink="false">https://shiftmag.dev/?p=9785</guid>

					<description><![CDATA[<p>At Devoxx UK, I spoke with Trisha Gee - author and one of the most recognized voices in the Java space - about what really happens when teams lean heavily on AI. Her take was far darker than the conference hype.</p>
<p>The post <a href="https://shiftmag.dev/trisha-gee-ai-wont-fix-your-broken-pipeline-it-will-break-it-faster-9785/">Trisha Gee: AI Won&#8217;t Fix Your Broken Pipeline &#8211; It Will Break It Faster</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></description>
										<content:encoded><![CDATA[<figure class="wp-block-post-featured-image"><img loading="lazy" decoding="async" width="1200" height="720" src="https://shiftmag.dev/wp-content/uploads/2026/05/tirsha-devoxx.jpg?x32039" class="attachment-post-thumbnail size-post-thumbnail wp-post-image" alt="" style="object-fit:cover;" srcset="https://shiftmag.dev/wp-content/uploads/2026/05/tirsha-devoxx.jpg 1200w, https://shiftmag.dev/wp-content/uploads/2026/05/tirsha-devoxx-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/05/tirsha-devoxx-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/05/tirsha-devoxx-768x461.jpg 768w" sizes="auto, (max-width: 1200px) 100vw, 1200px" /></figure>

<div class="wp-block-post-excerpt wp-block-hidden-desktop wp-block-hidden-mobile wp-block-hidden-tablet"><p class="wp-block-post-excerpt__excerpt">At Devoxx UK, I spoke with Trisha Gee &#8211; author and one of the most recognized voices in the Java space &#8211; about what really happens when teams lean heavily on AI. Her take was far darker than the conference hype. </p></div>


<p class="wp-block-paragraph"><strong>Trisha Gee</strong> has spent over two decades in software development, from startups to global enterprises &#8211; equally at home discussing DORA metrics and SPACE frameworks as business outcomes and organizational design.</p>



<p class="wp-block-paragraph">At <a href="https://www.devoxx.co.uk/" target="_blank" rel="noreferrer noopener">Devoxx UK</a>, she gave a talk about how <strong>software engineering principles stay the same regardless of what tooling era you are in</strong>. </p>



<p class="wp-block-paragraph">I wanted to understand what that means right now when AI is writing a significant portion of the code.</p>



<h2 class="wp-block-heading" id="ai-exposes-the-weakest-link-not-just-the-fastest-path">AI exposes the weakest link, not just the fastest path</h2>



<p class="wp-block-paragraph">Trisha frames AI as an amplifier, not a solution. When I asked what that looks like beyond demos, she put it simply: <strong>it exposes the problems that were already there</strong>, the ones you didn&#8217;t know you had.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">The most common thing I saw (I was working at Gradle, so we dealt with a lot of build tooling) was more code, more tests, and tests taking longer. The continuous delivery pipeline took a lot of pressure.</p>
</blockquote>



<p class="wp-block-paragraph">The broader pattern she describes is straightforward but easy to miss when you are excited about shipping faster. &#8220;Whichever part of your system is the weakest, it&#8217;s going to expose that part,&#8221; she said.</p>



<p class="wp-block-paragraph">Reframing it this way, while most conversations about AI adoption focus on what gets faster, Trisha highlights what deteriorates first.</p>



<h2 class="wp-block-heading" id="when-code-gets-cheap-everything-else-gets-expensive">When code gets cheap, everything else gets expensive</h2>



<p class="wp-block-paragraph">When I asked Trisha where teams should focus once code generation becomes cheap, her answer was <em>everywhere</em>.</p>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/05/crowd-devoxx-1024x614.jpg?x32039" alt="" class="wp-image-9877" srcset="https://shiftmag.dev/wp-content/uploads/2026/05/crowd-devoxx-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/05/crowd-devoxx-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/05/crowd-devoxx-768x461.jpg 768w, https://shiftmag.dev/wp-content/uploads/2026/05/crowd-devoxx.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">&nbsp;Photo: DevoxxUK / Flickr</figcaption></figure>



<p class="wp-block-paragraph">What she means is that optimizing the <strong>writing of code without understanding the surrounding system does not move the needle</strong>.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">It&#8217;s not about one thing which is going to fix one problem, it&#8217;s about really understanding the whole system, it&#8217;s about understanding even the whole organization, the whole enterprise. Where does IT and technology and software fit into that? What are you really trying to deliver? What is the business benefit?</p>
</blockquote>



<p class="wp-block-paragraph">She described this as <strong>working across two ends of the process</strong>. On the input side, teams need to get better at questioning requirements before writing anything. On the output side, they need to look at build pipelines, test parallelism, flaky tests, and DORA metrics.</p>



<p class="wp-block-paragraph">&#8220;If you can measure those things (your DORA metrics, build times, whether delivered requirements actually give users value) you can start to see which parts of the process are working and which need attention,&#8221; Trisha explained.</p>



<h2 class="wp-block-heading" id="measuring-the-wrong-things-optimizes-the-wrong-things">Measuring the wrong things optimizes the wrong things</h2>



<p class="wp-block-paragraph">She also makes a sharp point about measurement and optimization.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">If you measure lines of code for productivity, you&#8217;ll get more lines of code. But really productivity is not just about what we call these activity metrics. It&#8217;s not just lines of code. It&#8217;s not just pull requests, merges, features delivered.</p>
</blockquote>



<p class="wp-block-paragraph">The thing teams consistently miss is the full arc of delivery.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Developer experience and productivity is the whole piece. Did it get out to the user? Did it meet the user&#8217;s needs? Is the user paying for more of our stuff? Is the business getting what they need from what the developers are doing? What you&#8217;re measuring there impacts what you&#8217;re going to optimize.</p>
</blockquote>



<p class="wp-block-paragraph">That last line is worth sitting with. If your productivity metrics stop at pull requests merged, you are optimizing for pull requests merged.</p>



<h2 class="wp-block-heading" id="the-space-framework-and-why-three-metrics-beat-one"><br>The SPACE framework and why three metrics beat one</h2>



<p class="wp-block-paragraph">When I asked Trisha what teams should measure, she pointed to the SPACE framework. SPACE stands for <strong>satisfaction</strong>, <strong>performance</strong>, <strong>activity</strong>, <strong>communication and collaboration</strong>, and <strong>efficiency and flow</strong>.</p>



<p class="wp-block-paragraph">DORA metrics, which most teams are more familiar with, are a subset of it. Her recommendation is to <strong>pick metrics from three different dimensions</strong> rather than relying on a single category. The reasoning is that single-category metrics tend to be easy to game without improving anything real.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">So yes, you can write more code, but no, you didn&#8217;t do what the business wanted.</p>
</blockquote>



<figure class="wp-block-image size-large"><img loading="lazy" decoding="async" width="1024" height="614" src="https://shiftmag.dev/wp-content/uploads/2026/05/Trisha-Gee-1-1024x614.jpg?x32039" alt="" class="wp-image-9879" srcset="https://shiftmag.dev/wp-content/uploads/2026/05/Trisha-Gee-1-1024x614.jpg 1024w, https://shiftmag.dev/wp-content/uploads/2026/05/Trisha-Gee-1-300x180.jpg 300w, https://shiftmag.dev/wp-content/uploads/2026/05/Trisha-Gee-1-768x461.jpg 768w, https://shiftmag.dev/wp-content/uploads/2026/05/Trisha-Gee-1.jpg 1200w" sizes="auto, (max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Photo: Marin Pavelić</figcaption></figure>



<p class="wp-block-paragraph">She also brought up Fred Brooks and <strong>communication overhead</strong> as something the industry consistently underweights. The harder metrics to capture, like satisfaction and flow, are often more revealing than the activity metrics that dashboards make easy to track.</p>



<p class="wp-block-paragraph">The business outcomes she keeps returning to are specific: &#8220;You need to measure, did it do what you wanted it to do? Did it get out to the user in time? Did they start spending more money with us? Did it fix your retention problem?&#8221;</p>



<p class="wp-block-paragraph">Those are the things which matter much more to the business.</p>



<h2 class="wp-block-heading" id="what-to-fix-before-adopting-ai">What to fix before adopting AI</h2>



<p class="wp-block-paragraph">I wondered what teams need to get right before AI tooling can actually help them. Trisha&#8217;s first answer was essentially: stop adopting AI the way you have adopted everything else. </p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">We generally get requirements, write the code, chuck it out there, and then you&#8217;re kind of done. That&#8217;s not how it should work.</p>
</blockquote>



<p class="wp-block-paragraph">What she advocates for instead is <strong>applying the scientific method to engineering decisions</strong>, which sounds obvious but rarely happens in real life.</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">Have a hypothesis, do your investigation, measure the results, have a conclusion. Generally speaking, we have not been great at that in our industry.</p>
</blockquote>



<p class="wp-block-paragraph">Applied to AI adoption specifically, that means being precise about what you are actually trying to achieve. What are we trying to achieve with AI? Do we want to deliver more features more quickly to the customer or do we want to perhaps deliver higher quality features? Because those two things are not necessarily the same thing Trisha concluded.</p>



<p class="wp-block-paragraph">Therefore the practical instruction she gives is to <strong>run short experiments, measure one change at a time, and iterate</strong>. But have a hypothesis, figure out how to measure it, measure it, get feedback, and iterate over that.</p>



<figure class="wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio"><div class="wp-block-embed__wrapper">
<iframe loading="lazy" title="AI Won&amp;apos;t Fix What You Can&amp;apos;t Measure" width="500" height="281" src="https://www.youtube.com/embed/iH4UnTskOSM?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
</div></figure>



<p class="wp-block-paragraph"><br></p>
<p>The post <a href="https://shiftmag.dev/trisha-gee-ai-wont-fix-your-broken-pipeline-it-will-break-it-faster-9785/">Trisha Gee: AI Won&#8217;t Fix Your Broken Pipeline &#8211; It Will Break It Faster</a> appeared first on <a href="https://shiftmag.dev">ShiftMag</a>.</p>
]]></content:encoded>
					
		
		
			</item>
	</channel>
</rss>

<!--
Performance optimized by W3 Total Cache. Learn more: https://www.boldgrid.com/w3-total-cache/?utm_source=w3tc&utm_medium=footer_comment&utm_campaign=free_plugin

Page Caching using Disk: Enhanced 

Served from: shiftmag.dev @ 2026-11-16 22:36:17 by W3 Total Cache
-->