Cognichip Wants AI to Design the Chips That Power AI — And Just Raised $60M to Try
Intel CEO Backs Cognichip: The $93M race to automate semiconductor design.
The Big Idea: Can AI Design the Chips That Run AI?
Advanced silicon chips made the AI revolution possible. Now a bold new startup is asking whether AI can return the favor — by designing the next generation of chips itself. That is the audacious bet behind Cognichip, a deep learning company building an AI model that works alongside engineers as they design new computer chips. It is a problem the semiconductor industry has wrestled with for decades, and Cognichip thinks it finally has the tools to crack it.
Chip design is one of the most complex engineering disciplines on the planet. Advanced chips take three to five years to go from conception to mass production, with the design phase alone consuming up to two years before physical layout even begins. To put the scale in perspective: Nvidia's latest Blackwell GPU line contains 104 billion transistors — each one needing to be precisely placed and interconnected. The cost, the time, and the complexity are all immense.
"These systems have now become intelligent enough that by just guiding them and telling them what the result is that you want, it can actually produce beautiful code." — Faraj Aalaei, CEO & Founder, Cognichip
The Problem: Slow, Expensive, and Ripe for Disruption:
Cognichip CEO and founder Faraj Aalaei has spent his career watching the semiconductor industry struggle with an innovation bottleneck. In the time it takes to bring a new chip from concept to market, the competitive landscape can shift completely — rendering years of engineering effort and billions in investment obsolete before a single unit ships. Aalaei's mission is to bring the same kind of AI-powered productivity tools that have transformed software engineering into the semiconductor design space.
The promise is staggering in its ambition. Aalaei says Cognichip's technology can reduce the cost of chip development by more than 75% and cut the design timeline by more than half. If those numbers hold up under real-world conditions, it would be one of the most significant productivity leaps in semiconductor history — compressing years of work and hundreds of millions of dollars in engineering costs into a fraction of either.
The Funding: $60 Million and a Star-Studded Board:
Cognichip has just made its biggest statement yet that it is serious about this mission. The company, which emerged from stealth last year, announced Wednesday that it had raised $60 million in new funding led by Seligman Ventures. The round brings Cognichip's total fundraising to $93 million since its founding in 2024 — a remarkable sum for a company that has not yet publicly demonstrated a finished chip designed with its system.
The investor lineup carries serious weight. Intel CEO Lip-Bu Tan participated in the round and will be joining Cognichip's board — a notable endorsement from one of the most powerful figures in the global semiconductor industry. Umesh Padval, a managing partner at Seligman Ventures with four decades of investing experience, will also join the board. Their involvement signals confidence that Cognichip is tackling a real and urgent problem.
"If it's a super cycle for semiconductors and hardware, it's a super cycle for companies like [Cognichip]." — Umesh Padval, Managing Partner, Seligman Ventures
Padval was characteristically direct about the scale of the moment. He described the current flood of capital into AI infrastructure as the largest investment super cycle he has witnessed in 40 years of investing — a generational wave of capital chasing the hardware that will define the next era of computing.
The Technology: A Domain-Specific Model, Not a General LLM:
What sets Cognichip apart from simply plugging ChatGPT into a chip design workflow is its foundational approach. Rather than starting with a general-purpose large language model and fine-tuning it for semiconductor work, Cognichip built its own deep learning model trained specifically on chip design data from the ground up. The company believes that domain specificity is the key advantage — and the key challenge.
Getting that training data was no small feat. Unlike the software world, where developers share vast amounts of code openly and AI coding assistants can train on enormous public repositories, chip designers guard their intellectual property fiercely. The kind of open-source trove that powers tools like GitHub Copilot simply does not exist in the semiconductor world. Cognichip has had to build its own datasets, generate synthetic data, and license proprietary information from industry partners.
The company has also developed a privacy-preserving approach for chipmakers. Cognichip's procedures allow semiconductor companies to securely train its models on their own proprietary design data without ever exposing that data to the outside world — a critical capability for an industry that treats its designs as crown jewels.
The Demo: Engineering Students, Open Source Chips, and a Hackathon:
Where proprietary data isn't available, Cognichip has turned to open source alternatives to prove its model's capabilities. In one notable demonstration last year, the company invited electrical engineering students at San Jose State University to put its technology to the test in a hackathon format. The results were striking: student teams were able to use Cognichip's model to design fully functional CPUs based on the RISC-V open source chip architecture — a freely available design standard that anyone can build on.
The RISC-V hackathon was more than a marketing exercise. It demonstrated that even non-expert users — engineering students, not seasoned chip designers — could leverage Cognichip's AI model to produce real, working chip designs. If the technology can do that with students, the implication is clear: in the hands of professional engineers, its potential impact is significantly greater.
The Competition: A Well-Funded Race to Reinvent Chip Design:
Cognichip is not operating in a vacuum — the race to AI-power semiconductor design has attracted serious capital and serious competitors. The company faces competition from semiconductor design incumbents like Synopsys and Cadence Design Systems, both of which have been integrating AI capabilities into their established toolchains for years. But the more urgent competitive pressure may come from well-funded newcomers.
The startup landscape in this space is moving fast. ChipAgents closed a $74 million extended Series A in February, while Ricursive raised a $300 million Series A round in January — a staggering sum that underscores just how much investor conviction is flowing into AI-driven semiconductor design. Cognichip's $60 million raise keeps it competitive, but the arms race is clearly escalating.
Still, Cognichip cannot yet point to a new chip designed with its system — and has not disclosed any of the customers it says it has been collaborating with since September. That transparency gap is the honest asterisk on an otherwise compelling story. The technology promises are bold, the investor roster is impressive, and the problem being solved is real. But until Cognichip can point to a production chip designed with its AI — or name a customer willing to go on record — it remains a company with enormous potential still working to prove itself.
Conclusion: A Super Cycle Moment for the Chip Industry:
The timing for what Cognichip is attempting could not be more consequential. The AI boom has triggered an insatiable appetite for more powerful, more efficient chips — and the traditional methods of designing them are straining to keep up. Three-to-five year development cycles and billion-dollar price tags are not compatible with the pace at which AI is advancing. Something has to give.
Cognichip is betting that AI is the answer to its own hardware problem. If Faraj Aalaei and his team can deliver on even a fraction of their promises — cutting design timelines in half, slashing costs by 75%, and democratizing chip design with AI assistance — the impact on the semiconductor industry would be transformational.
With $93 million raised, Intel's CEO on the board, and a $60 million vote of confidence from Seligman Ventures, Cognichip has earned the right to try.
