Operando optical measurement
Track optical response across the substrate while potential and current are controlled, retaining the time and spatial information lost in a single electrode-wide signal.
Semiconductor process metrology
FabSight is developing operando optical metrology and AI-assisted analysis for semiconductor electrodeposition, starting with copper feature filling on patterned wafers.
Bulk signals can hide local failure.
Feature geometry, additives, transport and surface evolution interact during plating. FabSight is being built to reveal where those differences emerge across a patterned surface, including collective optical trajectories from repeated features smaller than the diffraction limit.
Technology
FabSight combines synchronized optical imaging, electrochemical control and physics-informed, AI-assisted analysis to turn patterned substrates into information-rich process experiments.
Track optical response across the substrate while potential and current are controlled, retaining the time and spatial information lost in a single electrode-wide signal.
Link image sequences to potential, current, waveform and process conditions so optical transitions can be interpreted against the electrochemical experiment.
Use dimensionality reduction, clustering and predictive models to interpret complex time- and potential-dependent optical trajectories, including spatially averaged signals from repeated sub-diffraction feature populations.
The FabSight workflow
The initial platform is designed for process development rather than production-line deployment. It is intended to help teams compare chemistries, process windows and patterned test structures while extracting reproducible optical trajectories that evolve with the deposition itself.
Applications
FabSight's first development focus is copper electrodeposition, with a platform architecture intended to extend to other metals and geometries as the measurement is validated.
Compare how additive chemistry and process conditions affect patterned feature populations during copper growth.
Evaluate spatial differences across larger recessed structures where geometry and transport can strongly influence filling behavior.
Generate time-resolved, spatial evidence for how suppressors, accelerators, levelers and other bath variables alter the deposition response.
Current stage
FabSight is at an early development stage. The current program is focused on building and validating an Alpha-1 instrument, separating useful interfacial and growth information from thickness and roughness effects, and developing AI-assisted methods to identify reproducible optical trajectories across patterned substrates.
Company
FabSight is an Oregon-based semiconductor metrology startup developing tools for observing and understanding electrodeposition on patterned substrates. The company combines electrochemistry, optical measurement and AI-assisted data analysis around a simple premise: process development is more informative when the measurement can reveal where, when and how different growth trajectories emerge.
Founder & Technical Lead
Electrochemist and surface-science researcher working across electrochemical interfaces, operando optical measurements, semiconductor electrodeposition and data-driven interpretation of dynamic process signals.
Work with FabSight
We're interested in conversations with semiconductor process-development teams, plating-chemistry suppliers, equipment developers and early technical partners exploring new ways to understand electrodeposition.