Key Takeaways
- • Agentrys closed a $24.5 million funding round comprising a $19.1 million oversubscribed seed led by Etna Labs and a $5.4 million pre-seed from MediaTek, targeting chip design automation and workflow expansion.
- • The platform exceeds 90% accuracy on NVIDIA's public CVDP verification benchmark, positioning Agentrys among the first to achieve this performance threshold using agent evolution techniques.
- • Founded by Mark Ren, a veteran of NVIDIA Research and IBM Research with nearly three decades of EDA and AI experience, including leadership on ChipNeMo, the first industrial large language model for chip design.
- • Agentrys is defining a new category—Agentic Design Automation (ADA)— that shifts chip-design automation from traditional task-focused EDA tools to self-improving intelligent systems that automate, learn from, and continuously optimize entire engineering workflows.
- • Capital will fund recruitment, agent-native tooling development, and customer expansion across verification and physical design, with engagements underway at major semiconductor companies and foundries.
Summary
Agentrys, a chip design automation startup, announced $24.5 million in total funding on August 26, 2026, marking a significant capital raise for an emerging category of AI-driven semiconductor design tools. The round comprised an oversubscribed $19.1 million seed led by venture capital firm Etna Labs, following an initial $5.4 million pre-seed round led by MediaTek, a major semiconductor company serving as Agentrys' first strategic backer. The funding targets three strategic priorities: recruiting engineering talent, developing agent-native tooling infrastructure, and scaling customer deployments across chip verification and physical design workflows.
The startup has achieved a notable technical milestone by exceeding 90% accuracy on NVIDIA's public CVDP (Circuit Verification and Design Platform) verification benchmark, positioning it among the first entrants to this performance threshold. This achievement, accomplished through proprietary agent evolution techniques, validates the core technical approach underpinning the platform's value proposition. The company is actively engaged with multiple major semiconductor companies and foundries, indicating early market traction in an industry where manual engineering and scarce expertise currently represent significant bottlenecks.
Agentrys is pioneering what it defines as Agentic Design Automation (ADA), a category distinction from traditional Electronic Design Automation (EDA) tools. Rather than automating discrete, isolated tasks within chip design workflows, ADA systems operate as self-improving agents capable of learning from experience, evolving their capabilities, and optimizing entire engineering workflows. This architectural shift addresses a fundamental pain point in the $800+ billion semiconductor industry: chip design remains dependent on scarce expert engineers and thousands of manual engineering hours per project, creating bottlenecks that impact time-to-market and design quality.
Founder and CEO Mark Ren brings extensive credibility to the venture, with nearly three decades of experience spanning Electronic Design Automation and artificial intelligence research. His background includes leadership roles at NVIDIA Research and IBM Research, most notably directing ChipNeMo, recognized as the first industrial-scale large language model purpose-built for chip design applications. This pedigree suggests deep domain expertise and industry relationships critical for enterprise adoption in semiconductor manufacturing.
The capital raise and technical validation position Agentrys at the intersection of two high-demand trends: the semiconductor industry's urgent need for design efficiency and the broader institutional shift toward agentic AI systems. Success in this category could address structural inefficiencies affecting chip delivery schedules and functional performance—outcomes material to semiconductor companies' capital allocation and operational metrics.
The startup has achieved a notable technical milestone by exceeding 90% accuracy on NVIDIA's public CVDP (Circuit Verification and Design Platform) verification benchmark, positioning it among the first entrants to this performance threshold. This achievement, accomplished through proprietary agent evolution techniques, validates the core technical approach underpinning the platform's value proposition. The company is actively engaged with multiple major semiconductor companies and foundries, indicating early market traction in an industry where manual engineering and scarce expertise currently represent significant bottlenecks.
Agentrys is pioneering what it defines as Agentic Design Automation (ADA), a category distinction from traditional Electronic Design Automation (EDA) tools. Rather than automating discrete, isolated tasks within chip design workflows, ADA systems operate as self-improving agents capable of learning from experience, evolving their capabilities, and optimizing entire engineering workflows. This architectural shift addresses a fundamental pain point in the $800+ billion semiconductor industry: chip design remains dependent on scarce expert engineers and thousands of manual engineering hours per project, creating bottlenecks that impact time-to-market and design quality.
Founder and CEO Mark Ren brings extensive credibility to the venture, with nearly three decades of experience spanning Electronic Design Automation and artificial intelligence research. His background includes leadership roles at NVIDIA Research and IBM Research, most notably directing ChipNeMo, recognized as the first industrial-scale large language model purpose-built for chip design applications. This pedigree suggests deep domain expertise and industry relationships critical for enterprise adoption in semiconductor manufacturing.
The capital raise and technical validation position Agentrys at the intersection of two high-demand trends: the semiconductor industry's urgent need for design efficiency and the broader institutional shift toward agentic AI systems. Success in this category could address structural inefficiencies affecting chip delivery schedules and functional performance—outcomes material to semiconductor companies' capital allocation and operational metrics.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.