Google Reorganizes Project Mariner Team Amid AI Code Agents Boom

Google is restructuring the team behind Project Mariner — its Gemini-based agent that can independently work in a browser and perform web-based tasks. The reorganization is taking place against the backdrop of a sharp industry-wide shift of interest toward AI agents for code writing.

Project Mariner emerged in late 2024 as one of the first public demonstrations of a new class of agents — those capable of controlling a browser like a human: clicking, filling out forms, following links. At the time, it seemed like the next big step.

But by 2025–2026, the main battleground has shifted from web navigation to code generation and editing.

In recent months, Silicon Valley has been experiencing another wave of hype — this time around agents capable of autonomously writing, testing, and deploying code. OpenAI, Anthropic, Microsoft, and a number of startups have invested significant resources precisely in this direction.

Under these circumstances, browser agents — impressive as they are technically — have taken a back seat in the priority system.

According to Wired, the reconfiguration of the Project Mariner team reflects a broader strategic reorientation of Google: resources are being reallocated to where competition is currently concentrated.

This does not mean the project is being shut down — rather, a shift in focus and pace of work.

History repeats itself: the AI industry moves in waves, and each new hype cycle redistributes talent and funding.

Today's flagship is coding agents. What the next breakthrough segment will be a year from now remains to be seen, but Google, it appears, prefers not to fall behind the current trend rather than betting solely on long-term technological niches.