Google has completed a talent acquisition of Mechanize, a San Francisco-based startup focused on improving AI models' software engineering abilities, according to a report from Business Insider on September 12. The move is part of Google's effort to close the gap with rivals Anthropic and OpenAI in AI coding.
Team moves to DeepMind
Mechanize co-founder Tamay Besiroglu updated his LinkedIn profile to show he joined Google DeepMind in August, and reports indicate that more than a dozen Mechanize engineers have also moved to the division. The financial terms were not officially disclosed, but Google was previously reported to be in talks for a deal worth up to $1.5 billion (about 10.1 billion yuan) to secure Mechanize's technology and talent.
Mechanize, which raised $9.1 million (about 61 million yuan) earlier this year at a $500 million (about 3.36 billion yuan) valuation, is expected to continue operating other business lines despite losing its founding team. The company is currently led by CEO Jiv Asadi.
Such deals, known as acqui-hires, allow tech giants to bring in key personnel and technology without the full antitrust scrutiny that formal mergers face from regulators like the U.S. Federal Trade Commission (FTC) or the UK's Competition and Markets Authority (CMA).
Why it matters for Gemini
Google previously used a similar approach to absorb the Windsurf team, whose former CEO Varun Mohan now leads Google's Antigravity AI IDE. The Mechanize hires are primarily focused on mid-training, aiming to strengthen model performance in software engineering and coding tasks—an area where some of Google's recent models have underperformed, according to the report.
Besiroglu also brings experience in AI model testing, having co-founded Epoch AI, a research group that builds AI benchmarks. The acquisition raises questions about whether Google can resolve delays with its Gemini Pro models. If not, Anthropic and OpenAI could extend their lead with upcoming models like GPT-6 Astra and Claude Fable.
According to IThome, the deal underscores the intensifying competition in AI development, where top talent is increasingly scarce and valuable.
