Subvolts take AI is now designing the chips that power AI, making hardware development faster and more accessible. Jan 14, 2026 - Developers
- Learn how AI is accelerating hardware design, potentially creating new tools and architectures for AI development.
- Business
- Understand how AI can democratize custom hardware, enabling companies to build specialized chips for their specific AI needs.
The founders previously built AI chips at Google and are now tackling the hardware bottleneck for AI development.
How Ricursive Intelligence’s Founders are Using AI to Shape The Future of Chip Design
Audio streams from Sequoia Capital. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Sequoia Capital's "Training Data" podcast features Anna Goldie and Azalia Mirhoseini, founders of Ricursive Intelligence. They discuss using AI to design computer chips, a process they've accelerated from years to hours. Their goal is to make custom chip creation accessible to any company, enabling AI to recursively design more powerful hardware and speed up AI development.
Building advanced computer chips usually takes years and lots of money. Anna Goldie and Azalia Mirhoseini explain how they use artificial intelligence to design chips much faster, cutting down the time from months to hours. This could let any company create its own specialized chips, speeding up AI progress.
This episode offers insight into the hardware bottleneck for AI advancement. Goldie and Mirhoseini, who previously worked on Google's TPUs, explain how AI can revolutionize chip design. Their work at Ricursive Intelligence aims to democratize custom silicon, potentially impacting how companies like Google, Microsoft, and Meta develop their AI infrastructure. It's for engineers, AI researchers, and business leaders focused on the future of AI hardware.
- HostSequoia Capital
- GuestAnna GoldieFounder, Ricursive Intelligence
- GuestAzalia MirhoseiniFounder, Ricursive Intelligence
- Guest and Title: Anna Goldie and Azalia Mirhoseini, Founders of Ricursive Intelligence
- Show: Training Data (Sequoia Capital)
- Length: 37 minutes
- Chip Design Improvement: Reduced chip floor planning from months to hours.
- Previous Work: Created AlphaChip at Google, using AI for four generations of TPUs.
Anna Goldie and Azalia Mirhoseini created AlphaChip at Google, using AI to design four generations of TPUs and reducing chip floor planning from months to hours. They explain how chip design has become the critical bottleneck for AI progress -- a process that typically takes years and costs hundreds of millions of dollars. Now at Ricursive Intelligence, they're enabling an evolution of the industry from “fabless” to "designless," where any company can create custom silicon with Ricursive Intelligence. Their vision: recursive self-improvement where AI designs more powerful chips, and faster, accelerating AI itself.
Hosted by Stephanie Zhan and Sonya Huang
- TPUs
- Azalia Mirhoseini
- Tensor Processing Units, specialized hardware chips designed by Google for AI tasks.
- floor planning
- Anna Goldie
- A stage in chip design that involves arranging the physical layout of components on the chip.
- fabless
- Azalia Mirhoseini
- A semiconductor company that designs chips but does not manufacture them.
- silicon
- Azalia Mirhoseini
- Refers to computer chips, made from silicon material.
1 h 08David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
WithDavid Silver, Lasse Espeholt
1 h 041 h 0552 minMaking Cities Awesome: Peregrine’s Nick Noone & Ben Rudolph
WithNick Noone, Ben Rudolph
34 min
44 minControlling AI Models from the Inside
WithAlizishaan Khatri
47 minHow Should AI Be Regulated? Use vs. Development
WithMartin Casado
Questions people ask
- What problem does Ricursive Intelligence solve in chip design?
- Chip design is a critical bottleneck for AI progress, typically taking years and costing millions. Ricursive Intelligence aims to make custom chip creation accessible to any company.
- How are AI and chip design connected in this episode?
- The founders used AI to design TPUs at Google and are now using AI to enable recursive self-improvement, where AI designs more powerful chips faster, accelerating AI itself.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Sequoia Capital.