Innatera Partners with Synopsys to Scale Brain-Inspired Edge AI
Delft, Monday 2 March 2026
Leveraging Synopsys simulation tools, Innatera is scaling its Pulsar chip—a brain-inspired processor delivering 500x lower energy consumption for efficient, always-on edge AI applications.
Strategic Alignment for Edge Intelligence
As of 1 March 2026, Innatera has formally selected Synopsys to provide the critical simulation and validation framework for its next-generation neuromorphic microcontrollers [3]. This partnership represents a significant step in the maturation of the Dutch semiconductor value chain, specifically within the realm of deeptech chip design. By adopting Synopsys’ industry-leading Electronic Design Automation (EDA) tools, Innatera aims to accelerate the development of processors capable of powering wearables, smart home devices, and digital twin industrial sensors [3]. The collaboration focuses on overcoming the inherent complexities of designing chips that mimic biological brain function, ensuring that European innovation remains at the forefront of the global edge AI market.
Overcoming Analog Design Challenges
To mitigate the risks associated with mixed-signal analog designs, Innatera is leveraging specific Synopsys solutions: PathFinder-SC and Totem [1]. The complexity of neuromorphic circuits requires a validation process that can manage noise coupling and maintain reliability without compromising the chip’s speed or energy efficiency [1]. PathFinder-SC is utilised to simulate ESD events at scale, identifying potential vulnerabilities and their root causes before the design moves to the manufacturing phase [1]. This capability is essential for ensuring that the final chips can withstand real-world electrostatic occurrences, a common failure point in delicate edge electronics [1].
Redefining Edge Performance with Pulsar
The immediate beneficiary of this enhanced design flow is Pulsar, Innatera’s flagship product and the world’s first commercial neuromorphic microcontroller [1]. The technical specifications of Pulsar highlight the immense potential of SNN-based architectures. The processor delivers up to 100x lower latency and 500x lower energy consumption compared to conventional AI processors [1]. These metrics are transformative for battery-powered devices, where energy constraints often limit the complexity of on-device processing. By processing data at the sensor edge rather than sending it to the cloud, Pulsar significantly improves data transfer speeds while maintaining ultra-low power operation [1].