PageSourceSearch

https://dynosaurlabs.com/assets/AiLeadershipPartTwo-BRP9a373.js

js dynosaurlabs.com collected 2026-10-07 16:41:32 UTC 10,474 bytes, 1 lines download raw bytes

1import{j as e,S as o}from"./index-C69_Iz-Z.js";import{R as l}from"./ReportPartLayout-DIbbW72f.js";import"./NewsletterSignup-D0t264pq.js";import"./input-CFS5hUlp.js";import"./ArticleShareControls-BEpij3pP.js";import"./linkedin-B0hOvVUG.js";import"./check-B1vLDlj5.js";import"./ArticleAttribution-uQkdIaQB.js";/* empty css                */const h="The State of Technical Leadership Hiring in AI.",i="Who actually runs AI engineering in Europe",n=`${h} Part II`,s="57% of Europe's verified AI leaders joined since January 2024. The class photo is still being taken.",r="https://dynosaurlabs.com/insights/who-runs-ai-engineering-europe",t="https://dynosaurlabs.com/insights/ai-leadership/v1/who-runs-ai-engineering-europe-share.webp",w=()=>{const a={"@context":"https://schema.org","@type":"Article",headline:`${n}: ${i}`,description:s,image:[t,"https://dynosaurlabs.com/insights/ai-leadership/chart-leadership-footprint.png"],datePublished:"2026-09-21",dateModified:"2026-09-21",inLanguage:"en-GB",isPartOf:{"@type":"CreativeWorkSeries",name:"The European AI Leadership Report 2026",position:2},author:{"@type":"Person",name:"Yves Greijn",url:"https://dynosaurlabs.com/authors/yves-greijn"},publisher:{"@type":"Organization",name:"Dynosaur Labs",url:"https://dynosaurlabs.com"},mainEntityOfPage:r,about:["AI engineering leadership","European AI companies","technical leadership hiring"]};return e.jsxs(e.Fragment,{children:[e.jsx(o,{title:`${n}: ${i} | Dynosaur Labs Insights`,description:s,canonical:r,ogImage:t,ogType:"article"}),e.jsx("script",{type:"application/ld+json",dangerouslySetInnerHTML:{__html:JSON.stringify(a).replace(/</g,"\\u003c")}}),e.jsxs(l,{part:2,title:n,deck:e.jsxs(e.Fragment,{children:[e.jsx("strong",{children:i}),". ",s]}),canonical:r,shareTitle:`${n}: ${i}`,publishedAt:"2026-09-21",modifiedAt:"2026-09-21",readingMinutes:7,toc:[{id:"class",label:"A class still settling in"},{id:"where",label:"Where they sit"},{id:"labs",label:"What the US labs run"},{id:"scaleups",label:"The scale-ups"},{id:"titles",label:"Flat titles"}],previous:{to:"/research/ai-leadership-hiring-europe-2026",label:"Part I: Why we built the dataset"},next:{to:"/insights/who-gets-hired-and-who-stays-ai-engineering-europe",title:"Who gets hired and who stays",summary:"Part III looks at who gets hired, whether a big-name background predicts time in seat and what separates the leaders who have been there longest."},children:[e.jsxs("div",{className:"journal-whitepaper-opening",children:[e.jsx("p",{className:"journal-whitepaper-opening-label",children:"The finding"}),e.jsx("p",{children:"We verified 110 technical leaders currently in seat at 20 AI companies across Europe. VPs, CTOs, Heads of Engineering and AI, Directors, Founding Engineers."}),e.jsx("p",{children:"Of those 110, 57 per cent joined their current company since January 2024. Nearly a third joined since January 2025. The median time in their current role is 14 months."}),e.jsx("p",{children:"Half of the people running engineering at Europe's top AI companies have been in the job for 14 months or less. These are not veterans of their organisations. They are recent arrivals still building the teams they lead."}),e.jsx("p",{children:"For anyone hiring against these companies, the number cuts both ways. The leadership benches look formidable on paper. Most of them are also newer in seat than the products they run, which means loyalty is shallow and the next 18 months will test who stays."})]}),e.jsx("h2",{id:"class",children:"A leadership class still settling into the job"}),e.jsxs("figure",{className:"journal-report-figure",children:[e.jsx("div",{className:"journal-report-figure-title",children:"When current leaders joined"}),e.jsxs("div",{className:"journal-report-stats",style:{"--cols":3},role:"img","aria-label":"57 per cent joined since January 2024, about one third joined since January 2025, median time in role is 14 months",children:[e.jsxs("div",{className:"is-primary",children:[e.jsx("strong",{children:"57%"}),e.jsx("span",{children:"Joined since January 2024"})]}),e.jsxs("div",{children:[e.jsx("strong",{children:"~1/3"}),e.jsx("span",{children:"Joined since January 2025"})]}),e.jsxs("div",{children:[e.jsx("strong",{children:"14"}),e.jsx("span",{children:"Median months in the current role"})]})]}),e.jsx("figcaption",{children:"110 verified technical leaders currently in seat at 20 AI companies across Europe."})]}),e.jsx("h2",{id:"where",children:"Where they sit and who they are"}),e.jsx("p",{children:"The United Kingdom hosts the largest share of Europe's AI technical leadership in our sample, roughly double Germany in second place, with the Netherlands and France following. For all the talk of Paris and Berlin as AI capitals, London remains where the leadership layer concentrates, with DeepMind's gravitational pull a large part of the story."}),e.jsx("p",{children:"On how they got there: 59 per cent were hired externally into their roles rather than promoted. And one more number worth holding onto: the median total career experience of these leaders is around 20 years. The people running AI engineering in Europe are not the twenty-something wunderkinder of the press coverage. They are industry veterans who crossed over."}),e.jsx("p",{children:"A note on the 14 months, because precision matters here. This is a snapshot figure: time served so far by the people currently in seat. It tells you how new this leadership class is. It does not tell you how long they will last; some will run for years yet. How long people actually last is a different measure and it gets its own post in a few weeks."}),e.jsx("h2",{id:"labs",children:"What the US labs actually run from Europe"}),e.jsx("p",{children:"Look at who actually holds leadership titles at the frontier labs' European operations and a pattern jumps out."}),e.jsx("p",{children:"The Europe-based engineering leadership we verified at OpenAI sits almost entirely in forward-deployed, customer success and deployment engineering functions. Anthropic's European footprint is smaller still, with a similar commercial orientation. These are field engineering roles: customer-facing, deployment-focused, close to revenue."}),e.jsx("p",{children:"There are two exceptions. Google DeepMind is the only US-anchored lab running genuine core technical leadership from Europe: research directors and senior engineering directors, with individual tenures in that group reaching levels nothing else in this dataset approaches. Mistral runs a real CTO and VP Engineering structure from Paris for the simple reason that Paris is home."}),e.jsx("p",{children:"Joining a US lab's European operation mostly means joining its go-to-market machine, however technical the title sounds. If you want to build the systems rather than deploy them, the honest European options are DeepMind, Mistral and the scale-up tier."}),e.jsx("h2",{id:"scaleups",children:"The scale-ups are where classical engineering leadership lives"}),e.jsx("p",{children:"At the labs in Europe the VP Engineering title barely exists. It is alive and well one tier down. DeepL, Helsing, Poolside, Wayve and Scale AI together account for the largest concentration of conventional engineering leadership in our sample. This matches what I see in searches. The scale-ups build their teams layer by layer. The labs work in a different way. You see it first in their job titles."}),e.jsxs("figure",{className:"journal-report-figure",children:[e.jsx("div",{className:"journal-report-figure-title",children:"The leadership footprint in Europe"}),e.jsxs("ol",{className:"journal-report-steps",children:[e.jsxs("li",{children:[e.jsx("em",{children:"01"}),e.jsx("strong",{children:"US labs in Europe: field and deployment leadership"}),e.jsx("span",{children:"OpenAI mostly forward-deployed, customer success and deployment engineering. Anthropic smaller with a similar commercial focus."})]}),e.jsxs("li",{children:[e.jsx("em",{children:"02"}),e.jsx("strong",{className:"is-primary",children:"The exceptions: technical leadership at home"}),e.jsx("span",{children:"Google DeepMind with research directors and senior engineering directors. Mistral with a CTO and VP Engineering in Paris."})]}),e.jsxs("li",{children:[e.jsx("em",{children:"03"}),e.jsx("strong",{children:"The scale-ups: classical engineering leadership"}),e.jsx("span",{children:"DeepL, Helsing, Poolside, Wayve and Scale AI hold the largest group of conventional engineering leaders in the sample."})]})]}),e.jsx("figcaption",{children:"110 verified technical leaders at 20 European AI companies."})]}),e.jsx("p",{children:"A handful of companies in the sample show no classical engineering leadership layer in Europe at all. Sometimes that is deliberate design. Sometimes it is simply a gap. From the outside you cannot tell which one you are looking at."}),e.jsx("h2",{id:"titles",children:"Flat titles are absorbing experienced leaders and the habit is spreading"}),e.jsx("p",{children:"This part surprised me. Next to the 110 leaders we also found hundreds of former CTOs, VPs and Directors. They now work under individual contributor titles. At Anthropic 14 of them carry the title Member of Technical Staff. OpenAI has 4 more. The one that made me look twice is Lovable. This European scale-up has 12 of them."}),e.jsx("p",{children:"I want to be careful with interpretation because two very different stories produce the same data. One story is compression: experienced leaders accepting IC titles as the price of admission to the frontier. The other is convention: these people may hold significant de facto leadership under deliberately flat titles. From the outside you cannot fully distinguish the two. To be honest the line is sometimes blurry from the inside as well."}),e.jsx("p",{children:"Either way the consequence for the market is identical. Titles at AI companies have stopped carrying reliable information about scope. If you compare candidates on titles alone, you will c
1ompare the wrong things."}),e.jsx("hr",{}),e.jsx("p",{children:e.jsx("em",{children:"A bit of context on who's writing this. I run Dynosaur Labs, where we help founders, CTOs and investors build the engineering organisations behind AI companies. Senior IC to C-suite, engineering and product. When something in these findings hits close to home and you want to talk it through, reach out via dynosaurlabs.com. First conversation is always free and never a pitch."})})]})]})};export{w as default};

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.