New AI Automates Custom Chip Design for Faster Scaling
Ricursive Intelligence aims to compress design timelines to one week to meet a projected 10,000x rise in compute demand.
Updated on Oct. 15, 2026 in Semiconductors
Live Poll
Do you trust AI to handle complex design tasks as effectively as human experts?
Founders Anna Goldie and Azalia Mirhoseini have launched Ricursive Intelligence, a startup developing AI to automate computer chip design. The company utilizes their AlphaChip technology to generate unconventional, non-traditional chip layouts by optimizing the entire processor at once.
Why it matters
The company aims to accelerate chip development to keep pace with a projected 10,000x increase in compute demand over the next five years. This approach seeks to bypass the traditional block-based partitioning that currently limits chip design efficiency.
AlphaChip optimizes entire chip layouts by abandoning traditional, restricted block-based partitioning. The AI learns from each iteration to produce non-traditional architectures, including circular arrangements, that diverge from established grid-based layouts.
The players
Anna Goldie
Co-founder of Ricursive Intelligence and developer of chip-design AI architectures.
Azalia Mirhoseini
Co-founder of Ricursive Intelligence specializing in automated hardware design systems.
Ricursive Intelligence
A startup focused on using artificial intelligence to automate and accelerate the design of computer chips.
The details
Ricursive Intelligence uses an AI system that iterates on processor floorplanning—the process of determining the physical location of components on a chip—to improve design outcomes. By avoiding traditional block-based partitioning, the system treats the chip as a single, holistic optimization problem. The underlying mechanism relies on machine learning models that assess physical connectivity and thermal constraints to produce high-density, non-traditional layouts like circular configurations.
Timeline
October 14, 2026: The founders presented the startup at TechCrunch Disrupt.
Next five years: Compute demand is projected to increase 10,000x.
The Tech Race
Ricursive Intelligence is entering a competitive landscape focused on AI-assisted hardware development. Its goal is to compete with existing manual and semi-automated design methodologies by significantly shortening production timelines to as little as one week.
The technology aims to shorten chip design timelines to one week, which could facilitate faster development of custom hardware for specialized fields. This transition could eventually accelerate innovation in AR, VR, space technology, and biological computing platforms.
The takeaway
The founders aim to scale chip design capabilities to meet an exponential rise in compute demand. Observers should track whether the firm announces production partnerships with major silicon manufacturers to validate the performance of these non-traditional layouts.
Further reading
For more on industry shifts in chip architecture, visit /tech/semiconductors/.
Live Poll
Do you trust AI to handle complex design tasks as effectively as human experts?
