ChipAgents, a Santa Clara startup building autonomous AI agents for semiconductor engineering, closed a $60 million Series A2 round on July 29, bringing its expanded Series A to $134 million. B Capital led the new tranche, joining existing investors Bessemer Venture Partners, Micron, MediaTek, Ericsson, and ScOp Venture Capital.
The funding marks ChipAgents’ third capital injection since October 2025. The company closed a $21 million Series A led by Bessemer last fall, followed by a $50 million Series A1 led by Matter Venture Partners in February 2026. Total funding now stands at $134 million in roughly nine months, according to TechFundingNews.
What ChipAgents Builds
The company’s platform deploys domain-specific AI agents that operate inside semiconductor design and verification workflows. Rather than suggesting code completions, ChipAgents’ agents read specifications, generate production-grade register-transfer level (RTL) code and testbenches, run simulations, analyze coverage, and debug failures across multi-step engineering tasks.
“We’re enabling the industry to move beyond AI assistants toward autonomous agents that can execute meaningful engineering work,” CEO and founder William Wang told Reuters.
Wang, a UC Santa Barbara professor who previously worked on Amazon Q at AWS, founded ChipAgents in 2024. The platform integrates with existing electronic design automation (EDA) tools rather than replacing them, a strategic choice given how deeply entrenched incumbent EDA vendors are in semiconductor production pipelines, as The Next Web noted.
Why Verification Attracts Agents
Verification can consume more than 70% of a chip design project’s timeline, according to TNW’s reporting. The process involves compiling generated code, running assertions, executing simulations, and measuring coverage. Each step produces measurable feedback that agents can evaluate programmatically, making it a natural fit for autonomous systems compared to open-ended generative tasks.
ChipAgents has built a root-cause analysis system that examines design files, testbenches, error logs, and waveform data simultaneously. Multiple agents investigate potential explanations in parallel, testing hypotheses before surfacing results. Unite.AI reported that customer Whalechip used the platform during a system-on-chip debugging project, where it identified four critical bugs and reduced individual analysis rounds from days to between 15 and 60 minutes.
Customer Traction and Revenue Growth
ChipAgents reports deployment at more than 120 semiconductor companies, up from 80 in February and 50 at its October 2025 Series A close. Micron and MediaTek are both investors and production customers.
“Semiconductor engineering teams face rising pressure to deliver more sophisticated designs in less time. ChipAgents is addressing that challenge by enabling engineers to automate advanced design and verification work,” Henry Huang, investment director at Micron, told TechFundingNews.
The company reported 6x annual recurring revenue growth in H1 2026. TechFundingNews flagged that this represents a sharp deceleration from the 140x year-over-year ARR growth cited at its February round and 50x at the October close. Larger revenue bases naturally compress growth multiples, but the magnitude of the drop warrants context beyond the press release framing.
The EDA Incumbents Question
ChipAgents enters a market long dominated by Synopsys and Cadence, both of which are building their own AI features on top of deeply embedded customer relationships. The Next Web framed the core risk plainly: “a startup promising to replace their tools has to prove its agents are reliable enough to trust with silicon that costs millions to fabricate.”
The company’s integration-first approach, working alongside existing EDA toolchains rather than against them, may help sidestep direct confrontation with incumbents. But as agentic capabilities mature, the line between complement and competitor will blur. Ambiq, an ultra-low-power semiconductor developer, has already expanded ChipAgents deployment from evaluation to multiple engineering teams, a trajectory that suggests production confidence is building among early adopters, per Unite.AI.