The reason behind Google’s decision to spend a whopping $2.4 billion to hire six people in a game-changing deal has been revealed.
The tech giant made the bold move during the industry’s AI race, hiring Varun Mohan, who is the co-founder and CEO of artificial intelligence coding startup Windsurf.
The deal also included the condition that Google will hire other employees from Windsurf.
A spokesperson for Google said: “We’re excited to welcome some top AI coding talent from Windsurf’s team to Google DeepMind to advance our work in agentic coding.
Advert
“We’re excited to continue bringing the benefits of Gemini to software developers everywhere.”
So why is Google splashing so much cash? It all comes down to the current talent war in AI.
The deal from Google came after it was revealed that Windsurf had been in talks with Sam Altman’s OpenAI, with an alleged $3 billion deal being on the table, according to CNBC.
Advert
Many people have taken to social media to share their own thoughts on the deal, with one user writing on Reddit: “Here's Why Google Bought Windsurf: They just hired the Windsurf founders Varun Mohan & Douglas Chen to help them build Google Deepmind so they can go against Meta's Super Intelligence Lab & Cursor's vibecoding tech.
“They also have a non-exclusive license to Windsurf's IP that lets Google integrate the tech into Gemini/Android Studio. (Gemini is about to be LIKE THAT)”
This prompted replies from a lot of others, with one person saying: “Google is slowly winning the AI arms race as Google has done many times before in other areas. They are not dumb.”
Another wrote: “Don't sleep on Zuckerberg. Meta is going to come strong too, there's a reason Facebook, Instagram & Whatsapp are still making billions.
Advert
.jpg)
“Plus they just did the same "reverse acqui-hire" as Google, but with a data labeling startup (Scale AI). So their AI models are going to be better than Gemini pretty soon…”
A third commented: “Waste of money if you ask me, application layer AI is just regular front end work with fancy prompts. I.e. designing an agent is not that hard. You just set up a json contract with the LLM and ask it to go through states, run commands, give it context about its surroundings and what to do.”
And a fourth added: “First gen agents weren’t that hard to design. They’re going to be working on second gen which will likely be far more complex, not just simple wrapper front end prompting.”