\n<\/p>\n
Etched,<\/a> the AI chip startup founded by three Harvard dropouts in 2022, has closed a $300 million Series C funding round at a $10.3 billion valuation, co-founder and COO Robert Wachen tells TechCrunch.\u00a0\u00a0<\/p>\n The round was led by Sequoia, with Andreessen Horowitz, SK Hynix, Jane Street, and Diffusion Capital also participating, along with other, earlier investors. Other backers of the company include names like Peter Thiel, Andrej Karpathy, Dylan Field, Amjad Masad, and more.\u00a0<\/p>\n Etched was previously valued at $5 billion in December when it raised a $500 million round, meaning it has doubled its valuation in about seven months. The company says this is the highest valuation ever for a Sequoia-led Series C. Last month, Etched announced that it had successfully manufactured its homegrown chips<\/a>, that its first full systems were being tested by clients, and it had already booked $1 billion worth of orders.\u00a0<\/p>\n Etched launched at a time when the idea of building a chip specifically for AI models<\/a> based on transformer technology (the architecture behind most modern AI systems, including ChatGPT and Claude) was considered<\/a> wild if not wacky. The company is still battling the perception that its products \u2014 which are sold as full systems, not just chips \u2014 involve chips designed to run only specific LLMs.\u00a0\u00a0<\/p>\n That\u2019s not the case, Wachen explains. The systems can run any AI model, including Mixture of Experts models like DeepSeek and Qwen \u2014 an architecture that splits tasks across specialized sub-models rather than relying on one large model \u2014 as well as non-transformer designs like Mamba, which is built on a different underlying architecture<\/a> known as a state-space model. (Interestingly, the idea of etching parts of a specific AI model directly into silicon to boost performance isn\u2019t considered far-fetched anymore. Google is reportedly pursuing the same concept with its Frozen v2 chip<\/a> for Gemini.)<\/p>\n Still, Etched\u2019s claim to fame today is that it designed two new components from scratch to speed up inference \u2014 the computing process that happens after a user submits a prompt.\u00a0\u00a0<\/p>\n \u201cInference is built in two stages,\u201d Wachen says, \u201cprefill and decode.\u201d The \u201cprefill phase\u201d involves understanding the prompt, including context. It\u2019s mathematically and compute-intensive. The \u201cdecode\u201d phase generates the output tokens (the actual answer the user sees). It requires less computation but needs massive amounts of memory.\u00a0<\/p>\n Etched created a prefill chip that operates \u201cdramatically\u201d faster, he promises, \u201cby running at a much lower voltage than any other AI chip. We call this low-voltage inference.\u201d Lower voltage generates less heat, which allows the chip to pack in more transistors.\u00a0\u00a0<\/p>\n For the decode process, Etched created a new type of memory and \u201cinterconnect technology that we call cluster-scale memory. It allows many chips to connect together and use a shared memory pool at a very, very fast, low latency,\u201d he says. The result, Etched promises, is high speeds but lower costs.<\/p>\n Because the startup was launched before most of the tech world (besides Nvidia) understood AI\u2019s specialized compute needs, the founders, CEO Gavin Uberti, Wachen, and Chris Zhu, have faced plenty of skeptics<\/a> who kept doubting even after the company announced that its first batch of silicon had been successfully manufactured by TSMC.<\/p>\n Much of that comes from how few people have had access to the systems. So far, access has been limited to investors and early customers. In fact, that\u2019s how Etched landed its list of famous investors in the first place \u2014 by showing them private demos in its office.<\/p>\n \u201cAndrej Karpathy from Anthropic, Noam Brown from OpenAI, Geoffrey Hinton, as well as all the investors in the funding round \u2014 these are all people who actually tried the hardware and are very excited about it,\u201d Wachen says.\u00a0\u00a0<\/p>\n Still, it\u2019s been a long, difficult road with more to go until the rack systems are mass produced and delivered. The trio famously dropped out of Harvard<\/a> to launch Etched, not knowing then how to raise cash (much less the loads of it they would need) or how to hire.\u00a0<\/p>\n \u201cWe had no idea how hard it was going to be,\u201d Wachen says. \u201cI think we still have to be humbled by what it will take to actually get to scale.\u201d\u00a0<\/p>\n Wachen remembers landing in the Bay Area after telling his parents he was leaving school to do a startup, with no office or apartment arranged. He slept on the floor of a friend\u2019s unfurnished house.\u00a0<\/p>\n \u201cI remember staying in my friend\u2019s house that they were about to sell, using a towel as a blanket,\u201d he laughs.\u00a0The founders eventually set up the servers they needed to run the chip-design tools in the garage of an early employee and \u201cevery time it needed to be rebooted, he would call his wife, and she would go and hit the reboot button.\u201d\u00a0<\/p>\n Today, there are 400 people bustling in an office, and Etched operates a 2 megawatt data center there. And it just opened a new 80,000 square-foot, 10MW facility in Milpitas, down the road from its main San Jose office. <\/p>\n \u201cWe\u2019re running tokens in our lab today, working with some of the largest AI companies in the world,\u201d he says.\u00a0<\/p>\n Wachen also has a blanket now, and a mattress \u201cand a pillow even. Multiple pillows,\u201d he jokes.\u00a0\u00a0<\/p>\n More importantly, he and his co-founders never let the doubters stop them. \u201cIt\u2019s come a long way. It\u2019s a very, very different world. But I think, when you really think something\u2019s possible, and you just work at it for a long time, you can do it.\u201d<\/p>\n Note: This story was updated to include a reference to the new 10MW facility.<\/em><\/p>\n<\/div>\n When you purchase through links in our articles, we may earn a small commission<\/a>. This doesn\u2019t affect our editorial independence.<\/em><\/p>\n