TechCrunch: Cognichip raises $60M to use AI for chip design

On April 1, 2026, TechCrunch reported that Cognichip emerged from stealth with a $60 million funding round led by Seligman Ventures, with Intel CEO Lip-Bu Tan joining the board. The company is building a deep learning model that works alongside engineers in the semiconductor design process — not replacing them, but handling the computationally intensive parts of translating intent into a manufacturable chip design.

What Cognichip does

Semiconductor design is one of the most expensive and time-consuming processes in hardware development. A single chip can take years to design, test, and validate before it goes near a fabrication plant. Cognichip’s approach is to train models on the constraints and patterns of chip design — layer configurations, routing, power distribution — so that AI can handle the mechanical optimization that currently requires highly specialized engineers and months of iteration.

The company claims its technology can reduce chip development costs by more than 75% and cut development timelines by more than half. These are significant claims that will require validation at production scale, but even a fraction of those gains would be meaningful to the hardware industry.

Why it matters for design and product teams

Most UX and product designers work at a distance from semiconductor development. But as AI capabilities have pushed into hardware — purpose-built inference chips, edge computing, embedded systems in consumer devices — the boundary between software product design and hardware constraints has become more relevant.

For product managers and designers working on AI-native products where inference cost and latency matter, faster and cheaper chip iteration means more room to experiment with hardware configurations. It also points to a broader pattern: AI is being applied not just to outputs that designers work on, but to the infrastructure that those outputs eventually run on.

The Cognichip raise is worth watching as a signal that investment is moving into the hardware layer of the AI stack, which has implications for what becomes technically feasible to design against in the next few years.