SirenOpt

SirenOpt is a deep-tech company founded in 2022 as a spin-out of the UC Berkeley Chemical Engineering Department. From its base in Berkeley, California, it develops non-destructive, real-time materials characterisation and classification systems built on cold atmospheric plasma, machine learning and predictive analytics. The company's work sits at the intersection of physics and data: plasma is used as a probe, and physics-informed machine-learning models convert the resulting measurements into material fingerprints for industrial customers.

Its principal product, PlasmaSens, uses cold atmospheric plasma to generate chemical, electrical and thermal data, which are processed by physics-informed ML algorithms. A single measurement yields over 200,000 raw data elements across three modalities: optical emission spectrum, thermal response and voltage/current signals. The technical stack spans optical emission spectroscopy, thermal analysis, electrical signal analysis and defect detection, applied to advanced manufacturing analytics.

The platform is delivered in several forms: benchtop characterisation systems for laboratory analysis, robot-arm mounted inspection tools for piece-to-piece checks in production, and roll-to-roll characterisation solutions for continuous production lines. Target industries include semiconductors, batteries, aerospace, electronics and advanced manufacturing, with a particular focus on high-performance micro- and nano-materials. The company's stated emphasis is on sustainable production and non-destructive testing.