What Ricursive Intelligence does

Ricursive Intelligence is a frontier AI lab founded in September 2025 by Dr. Anna Goldie and Dr. Azalia Mirhoseini, the co-creators of AlphaChip (published in Nature 2021) at Google DeepMind. The company is developing a self-improving AI system that applies reinforcement learning, graph neural networks, and large language models to automate end-to-end semiconductor chip design, with the goal of creating a recursive feedback loop in which AI models design next-generation chips that in turn enable more capable AI models. Its technical stack includes AlphaChip (deep reinforcement learning for chip placement, deployed across four generations of Google TPUs and adopted by MediaTek), RL-CCD (DAC Best Paper 2023), Insta (DAC Best Paper 2025), and C3PO (ASP-DAC Best Paper 2026). The company claims its approach can compress chip design cycles from years to hours or weeks and reduce chip power consumption by up to 25% versus traditional Manhattan grid methods.

The company's commercial roadmap proceeds in three phases: first, accelerating portions of chip design for third parties; second, combining all stages of chip design into an end-to-end AI model that outputs GDSII files directly; and third, co-evolving frontier AI models with custom-designed hardware, including a proprietary silicon substrate layer. The go-to-market is enterprise sales-led, targeting hyperscalers, AI labs, and large companies with at-scale AI workloads that lack in-house chip design expertise. Ricursive explicitly does not position itself as an EDA competitor to Synopsys or Cadence, instead framing its offering as a shift from 'fabless' to 'designless.' Revenue mechanics are expected to combine subscription, usage-based compute charges, and co-development agreements with hyperscalers building custom silicon.

The company is headquartered in Palo Alto, California, with a small team of fewer than ten employees as of January 2026 and is actively hiring across engineering, research, and security functions. It has raised $335 million in total across a $35M seed round in December 2025 (led by Sequoia Capital at a $750M valuation) and a $300M Series A in January 2026 (led by Lightspeed Venture Partners at a $4B post-money valuation), with NVentures (NVIDIA), DST Global, Felicis, 49 Palms, and Radical AI participating. Pricing has not been publicly disclosed, and the product is targeted for commercial launch in 2026.

Name

Ricursive Intelligence

Legal name

Ricursive Intelligence

Website

https://ricursive.com

Company type

Private

Founded year

2025

Operating status

Operating

Headcount range

1–10 employees

Short description

Ricursive Intelligence is a Palo Alto-based frontier AI lab, founded in 2025 by the co-creators of AlphaChip, applying reinforcement learning, graph neural networks, and large language models to automate end-to-end semiconductor chip design for hyperscalers, AI labs, and companies lacking in-house chip design expertise.

Ownership category

akta.pro rank

Product category

Electronic Design Automation (EDA)

akta.pro primary industry

AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs) (HDAEANAJ)

Keywords

• AI chip design
• Semiconductor design automation
• Electronic design automation
• Chip layout optimization
• AI hardware co-design

Headquarters

HQ city

Palo Alto

HQ country

United States

HQ region

North America

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Infrastructure, Operations, Marketing or Sales

Revenue model

1. AI-Powered Semiconductor Design Services: The company plans to offer AI-driven chip design as a service to companies with at-scale workloads lacking in-house chip design expertise. Revenue is generated through design service contracts, compute infrastructure fees, and potentially licensing of AI-generated design outputs (GDSII files). The company is not positioning itself as an EDA company and will not compete with existing EDA toolchains, suggesting a services/platform model rather than traditional software licensing.
2. Full-Stack Semiconductor Design Platform: In later phases, the company plans to offer a full-stack platform for semiconductor design, potentially shifting the industry from 'fabless' to 'designless.' Revenue could be generated through platform subscriptions, usage-based compute charges for AI model training, and co-development agreements with hyperscalers building custom silicon.

Go-to-market motion2 records

Distribution channels1 record

Marketing channels4 records

Core offering

Ricursive Intelligence is a frontier AI lab developing an AI-driven chip design platform that combines reinforcement learning, graph neural networks, and large language models into a unified system to automate end-to-end semiconductor design. The platform creates a recursive feedback loop where AI models design next-generation chips that in turn accelerate future AI model training, targeting companies with at-scale AI workloads that lack in-house chip design expertise.

Product overview

Ricursive Intelligence operates as a frontier AI lab offering a full-stack semiconductor design platform that uses AI to automate chip design, shifting the industry from 'fabless' to 'designless.' The platform combines their research portfolio — AlphaChip (Nature 2021, RL-based chip placement), RL-CCD (DAC Best Paper 2023), Insta (DAC Best Paper 2025), and C3PO (ASP-DAC Best Paper 2026) — into a unified system. The company is developing a three-phase rollout: accelerating portions of chip design for third parties, combining all stages into an end-to-end model outputting GDSII files, and ultimately co-evolving frontier AI models with custom hardware including a proprietary silicon substrate layer. They are not positioning as an EDA company and will not compete with established EDA toolchains.

Differentiator

Problem solved

Functional benefit

Products and services

• AI Chip Design Platform

Quantifiable outcome

• Reduces chip design cycles from years to weeks or hours
• +4 more outcomes

Named customers2 records

Segments3 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability3 records

Feature7 records

Scale indicators6 records

Recent moves5 records

Expansion highlights6 records

Broad incumbents

• Siemens EDA: Third major EDA player (formerly Mentor Graphics) offering chip design and verification tools. Relevant as another broad incumbent with deep customer relationships and the resources to add AI-driven design automation features.
• Google DeepMind: Origin of Ricursive's founders and the AlphaChip technology; DeepMind continues internal AI-for-chip-design research for Google's TPU roadmap. Comparable as a research organization with proprietary AI-driven chip design capabilities, though it is not commercialized as a standalone product.
• NVIDIA: Dominant AI accelerator vendor and investor in Ricursive via NVentures. Overlaps with Ricursive's mission of advancing AI through better hardware, and represents both a potential strategic partner for AI silicon co-design and a customer/competitor for custom AI accelerators.
• Synopsys: Largest EDA software vendor with established chip design toolchains (Design Compiler, IC Compiler) and an emerging AI suite (DSO.ai). While Ricursive explicitly states it does not compete with EDA toolchains, Synopsys is the closest established comparable in semiconductor design automation and could move into Ricursive's AI-native design space.
• Cadence Design Systems: Second major EDA vendor offering chip design and verification tools, with growing AI/ML capabilities (Cerebrus, Virtuoso). Directly comparable as a broad incumbent in chip design automation; potential competitor if it expands AI-driven chip design beyond existing toolchains.

Emerging players

• Lightmatter: AI hardware company building photonic processors and interconnect for AI datacenters. Comparable as a custom AI silicon startup in the broader compute infrastructure ecosystem that Ricursive targets.
• SambaNova Systems: Custom AI chip company designing its own accelerators (RDU) for enterprise and government AI workloads. Comparable as a custom AI silicon player with potential customer/competitor overlap with Ricursive's design automation platform.
• Groq: Custom AI inference chip company building LPU accelerators for low-latency inference. Comparable as a custom AI silicon player that designs and ships its own chips, illustrating the broader custom-silicon market Ricursive aims to enable.
• Cerebras Systems: Custom AI accelerator company building wafer-scale chips for AI training and inference. Comparable as a custom AI silicon player that could be a Ricursive customer (if it adopts external design automation) or competitor (if it expands into design services for hyperscalers).
• Celestial AI: AI hardware company developing silicon photonics-based memory fabric and accelerators for AI workloads. Comparable as another custom AI silicon startup in the broader ecosystem where Ricursive's design automation platform could become relevant.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks5 records

Key highlights6 records

Customer concentration

• Xx.com/RicursiveAI
• LinkedInlinkedin.com/company/ricursive-intelligence

Revenue estimate

Valuation estimate

Number of profiles

Profiles2 records

Funding overview

Funding rounds2 records

Investors10 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Ricursive Intelligence