Skip to content
Martin Splitt

Martin Splitt

Developer Relations Engineer, Google Search Relations

Google

At a Glance

Role
Developer Relations Engineer, Google Search Relations, Zurich
At Google since
2018
Before Google
Head of Engineering at Archilogic
Known for
JavaScript SEO, rendering, and Search Central Live events
Martin Splitt is a software engineer who works on Google's Search Relations team in Zurich, Switzerland. His specialty is how search engines render, crawl, and index sites built with JavaScript, and he spends much of the year on stage at Google's Search Central Live events and on the Search Off the Record podcast. In 2026 his questions turned increasingly to AI Overviews, AI Mode, and the files people build for AI crawlers.

From Archilogic to Google Zurich

His background is engineering. He worked across JavaScript, DevOps, Python, and Scala and was Head of Engineering at Archilogic AG before joining Google, as his conference bio notes. At Google he became the team's specialist on JavaScript rendering and web performance, the person site owners ask when a framework-built page fails to show up in Search.

His title on the Google Search Central Blog is the informal "Googlebot whisperer," and his posts there cover crawling, robots.txt, and content delivery networks, along with the event announcements that send him around the world.

Search Central Live on Three Continents

In 2026 he announced Search Central Live stops in Toronto in April, Bogota and Ciudad de México in October, and Zurich in December. At the Toronto event he gave the opening talk on how Search works, walking through the life of a URL from discovered to crawled, indexed, and served, and reminding the audience that Google does not index every page at all times, as a detailed recap from the event describes.

The same event made AI the main subject. Speakers explained query fan-out, in which one question is split into many sub-queries, and Google's position that the principles behind classic search also apply to AI Overviews and AI Mode. Our GEO and AEO industry statistics track how many agencies now sell separate AI search services anyway.

The Case Against Markdown Pages for AI

In June 2026 he and John Mueller used an episode of Search Off the Record to push back on a popular AI SEO tactic: serving stripped-down Markdown versions of pages to AI crawlers. He granted the appeal. “And I think that’s also why people think it’s good for LLMs, because you have less stuff, less tokens,” he said.

Then he explained the cost. HTML carries links, navigation, and headers that show how a page connects to the rest of a site. “HTML with all the links and navigation and the headers and all that kind of stuff that kind of gets stripped out in the Markdown files that make the website are important to understand the structure and how this connects to the rest of the site,” he said. Losing that, he added, is probably not good for crawling and discovery. Our AI crawler blocking statistics show how sites are handling those same crawlers.

Asking Google's Engineers About AI

In May 2026 he hosted Nikola Todorovic, director of software engineering at Google Search, for an episode on how AI is changing Search, covering the move from classic results to AI Overviews and AI Mode, how Google tests launches, and why queries are getting longer and more conversational. He framed the question the web keeps asking: how the ecosystem keeps thriving now that AI features are part of Search.

For an engineer who made his name explaining why a JavaScript page did not render, the AI era is the same job at a larger scale: telling site owners what Google's systems can read, and why the plain structure of the web still counts.

Links

← All people

Sources