Cognichip emerges from stealth with the goal of using generative AI to develop new chips

Chips are an important component of the AI ​​industry. However new chips do not hit the market at the same speed as new AI models and products.

Cognichip There is a high target of creating a foundational AI model that can help bring new chips to the market quickly.

San Francisco-based cognichip is working to create a physicist-undefined foundational AI model that can use semiconductor agencies to speed up the development process of new chips. The company calls this method “artificial chip detective” and hopes that it can help increase the time of chip production by 50% and reduce related expenditures.

This ambitious concept comes from the senior Faraz Alai of the semiconductor industry, who worked in various organizations including Fujitsu Network Communication and Sentilium Communication.

Alai told TechCrunch that the original story of his company began in the 21st. At that time, Aley was a member of the Silicon Valley Leadership Group, which was often seen talking about their industry ill.

He was concerned about what was happening in the semiconductor industry. He said that he had given a presentation to the group about the Stark Drop in Venture Capital investment for investment in semiconductor companies, which reached 20 agreements in one year in 20 years, and only one or two in one year.

“I originally warned the other CEO that it couldn’t be good for us,” Alai said. “It may not be good for the semiconductor industry in America and what we need to do is to change it fundamentally about it. If that trend continues we are going to lose our competition. We’re going to lose the strength to bring new ideas to the table.”

Considering how long it takes to get to the market, these companies are not surprised that these companies are not attracting zealous investors, he said. Then he sat in this concept for almost a decade.

He went out to find Candu Venture In 20 2016, and through that fund, increase in AI startups can be seen. When he realized that the progress in the generator AI reached a point where it could be used to support some of the existing challenges in the semiconductor industry, he decided to launch the cognichip in 2024.

Cognicip has been working on stealth since then and AI experts from places like Stanford, Google and MIT have created a team so that the building can start. ALAE said that the model would take at least a few years to create a “final performance” but said companies should be able to assist before reaching this goal.

“When we arrive at this stage, this artificial chip intelligence agency, we will create a system that can actually act as expert engineer,” Alai said. “Once we have achieved this view, you can actually do the same thing with a fraction of humans and in a much less time.”

Cognicip is now deriving from stealth with $ 33 million in a round of $ 33 million in a round led by Lux Capital and Mafield with the participation of FPV and Cando Ventures.

Managing Partner Navin Chaddha, the managing partner of the Mepield, told TechCrunch that when he was introduced to Alai, he felt that they were “cut from the same clothes”. Most of the work in the semicircular industry is still being done by humans; He said that he thought that the time was correct to bring AI to the mixture.

“This is a major pain, and the solution that this company will provide will be a pain killer and not vitamin for the semiconductor industry,” said Chaddha. “If your man doesn’t work then Ii can do it where there is a deficit of talent? Number one, great party, second, [they are] A huge, trillion dollar to solve an original pain point in the industry. “

ALEE said he would hope that Cognicip would be able to help access to Democratizing chips so that more semiconductor agencies could start and invest in land. Simple access means small companies can also create more specific chips for specialized or small models, he said.

All of this will depend on when, or if the company can reach artificial chip intelligence.

“What we are doing is not a slightly extended change,” Aiye said. “We’re not building a [electronic design automation] Tools, we are not trying to tweet the process a bit. We’re trying to set a new goal for our industry and bring some big changes. “