Google cofounder Sergey Brin is taking direct action to address Google’s lagging position in the AI coding race. According to a report from The Information, Brin has assembled a specialized “strike team” within DeepMind to close the gap between Gemini and Anthropic’s Claude — which has become the preferred tool for software developers worldwide.
In a recent memo to DeepMind staff, Brin issued a stark assessment: “To win the final sprint, we must urgently bridge the gap in agentic execution and turn our models into primary developers.” The memo signals a significant shift in priorities for Google’s AI division, which has historically focused on general-purpose AI capabilities.
Claude’s Rise in the Developer Community
Anthropic’s Claude — particularly its Claude Code variant — has emerged as the de facto standard for AI-powered software development. The tool’s ability to understand complex codebases, execute multi-step coding tasks, and integrate seamlessly into developer workflows has created a formidable competitor to both OpenAI and Google.
“DeepMind engineers use Claude as a daily tool. Most of the rest of Google does not,” wrote former Google engineer Steve Yegge in a widely shared post on X. “Non-DeepMind engineers get pushed onto internal Gemini variants. This is not a picture of an engineering org that is fine.”
The discrepancy has become a point of internal concern. While DeepMind researchers have adopted Claude for their work, other parts of Google remain locked into using Gemini-based internal tools — a situation that some employees describe as counterproductive.
The Strike Team’s Mission
The newly formed strike team is tasked with a singular focus: transforming Gemini into a tool that can match or exceed Claude’s coding capabilities. The team will work on improving “agentic execution” — the ability of AI models to autonomously complete complex, multi-step development tasks without constant human intervention.
This is not merely about improving benchmark scores. The goal, according to Brin’s memo, is to create an AI system that can function as a “primary developer” — capable of understanding existing codebases, making appropriate modifications, and handling entire development workflows with minimal guidance.
The urgency reflects the competitive landscape. With Claude establishing itself as the coding assistant of choice for startups and enterprises alike, Google risks losing developer mindshare to a rival that was founded just years ago.
Implications for the AI Industry
The formation of the strike team highlights several key trends in the AI industry:
Vertical specialization wins: Claude’s success in coding demonstrates that domain-specific excellence can trump general capability. Google’s response acknowledges this reality by creating a dedicated team rather than hoping general improvements will trickle down.
Founders take notice: Sergey Brin’s direct involvement illustrates how critical AI has become to Google’s future. The cofounder’s active role in steering DeepMind’s direction signals that leadership views the Claude threat as existential.
Agentic AI is the next battleground: The focus on “agentic execution” confirms that the industry is moving beyond chat interfaces toward AI systems that can act autonomously. Whoever masters agentic AI will likely dominate the developer tools market.
For Google, the strike team represents a bet that it can close the gap quickly. For Anthropic, it is validation that its focused approach to AI development is paying off. And for the broader developer community, the competition promises ever-improving tools — even if the internal dynamics at Google remain messy.