Mach42 has raised £7 million, approximately $9.5 million, in a pre-Series A funding round to expand its artificial intelligence technology for analog semiconductor simulation. The financing was led by IP Group, with participation from existing investors BGF and Foresight Group. The Oxford-based company said the capital will support team growth, broader commercial adoption, and continued development of simulation tools designed for increasingly complex chip engineering workflows.
Addressing an Analog Verification Bottleneck
Semiconductor manufacturers are under pressure to develop more powerful and energy-efficient chips while shortening development cycles and controlling computing costs. Analog circuit verification remains a significant bottleneck because engineers must test whether designs will operate accurately and reliably before committing them to expensive manufacturing processes. Conventional simulation approaches can struggle with the scale and complexity of modern designs, creating demand for tools that increase testing coverage without sacrificing precision.
AI-Powered Simulation Technology
Mach42 applies advanced machine learning to create high-accuracy surrogate models that reproduce the behavior of complex analog and mixed-signal systems. These models are intended to work alongside established SPICE simulators, enabling engineering teams to perform simulations faster and explore substantially more operating conditions. According to the company, its platform can deliver up to 100 times greater verification coverage while reducing development time and computational requirements.
Focus on Power Management Devices
The company is initially concentrating its commercial efforts on power management devices, which regulate the delivery and use of electrical power across electronic systems. Efficient verification in this segment can help chip developers improve product quality, identify design problems earlier, and reduce costly delays before manufacturing. Mach42 believes successful adoption in power management could establish a pathway into other categories across the broader analog semiconductor market.
Supporting Agentic Design Workflows
Mach42 is positioning its technology for a semiconductor industry increasingly exploring agentic design processes, in which artificial intelligence systems can execute and coordinate parts of engineering workflows. Such systems require rapid access to dependable simulation results before they can make useful design decisions or evaluate possible circuit changes. By accelerating physics-accurate simulation, Mach42 aims to provide infrastructure that allows AI-driven tools to operate within established electronic design automation environments.
Expansion and Commercial Adoption
The new funding will be used to expand Mach42’s engineering and business development teams as the company works to increase access to its Discovery Platform. Its technology can generate surrogate models from relatively limited data and export them in formats including Verilog-A, SystemVerilog, and C or C++, supporting integration with commonly used engineering tools. This compatibility is intended to reduce adoption barriers for semiconductor companies that rely on existing simulators and verification processes.
Investor and Leadership Perspective
IP Group partner Lee Thornton said increasingly complex semiconductor designs are forcing engineering teams to shorten development cycles without compromising accuracy. Mach42 Chair Tim Haynes said the company’s tools are designed to complement established SPICE simulators as agentic design flows create greater demand for computing capacity. Chief Operating Officer Paul Neil added that the funding would help the company expand its platform while advancing its goal of making fast, physics-accurate circuit simulation available to engineering teams worldwide.
Mach42’s pre-Series A round gives the company additional resources to commercialize machine-learning technology aimed at one of semiconductor design’s most persistent technical constraints. Its strategy combines faster analog simulation, compatibility with existing engineering environments, and preparation for more automated and agentic chip-development processes. The company’s progress will depend on demonstrating that its models can maintain the accuracy, reliability, and scalability required by semiconductor teams operating under demanding production conditions.