Liang Wenfeng has grown DeepSeek into a $60 billion AI powerhouse in just three years, all while rejecting the typical Silicon Valley grind. According to a transcript from a nearly four-hour meeting with his employees, obtained by Tencent Technology, Liang emphasized a management style free of KPIs, overtime, and the usual performance posturing.
"We generally don’t work overtime," he said, which is striking for a company that has amassed about 173 million downloads since its public debut in January 2025. DeepSeek’s rise is remarkable given its relatively small team of around 160 employees, especially when compared to industry giants like OpenAI, which employs thousands. The company’s success has been fueled by open-source development and significantly lower AI training costs.
From Startup to AI Challenger
Founded in July 2023 in Hangzhou, DeepSeek swiftly moved from obscurity to posing a serious challenge to ChatGPT’s dominance in consumer AI. Its valuation skyrocketed from a $10 billion projection in April 2026 to between $45 and $52 billion by mid-2026, closing or nearing the close of a funding round exceeding $7.4 billion. Liang initially funded the company through his hedge fund, High-Flyer, but the latest funding round marked a pivot toward external investors, including state-linked stakeholders. This move highlights DeepSeek’s growing role as China’s leading contender against American AI dominance, especially amid semiconductor export restrictions limiting access to advanced chips.
Even though DeepSeek operates completely outside of the crypto sphere with no tokens or blockchain products, its impact has rippled through markets. The launch of DeepSeek’s R1 model in early 2025 shook up traditional tech stocks with AI exposure and sparked speculative surges in Solana-based meme tokens inspired by the company.
This content is informational and should not be considered financial advice.
